water

Article
Flood Forecasting by Using Machine Learning: A Study
Leveraging Historic Climatic Records of Bangladesh
Adel Rajab 1 , Hira Farman 2,3 , Noman Islam 2,3 , Darakhshan Syed 4 , M. A. Elmagzoub 5 , Asadullah Shaikh 6 ,
Muhammad Akram 1 and Mesfer Alrizq 6, *

1 Department of Computer Science, College of Computer Science and Information Systems, Najran University,
Najran 61441, Saudi Arabia; [email protected] (A.R.); [email protected] (M.A.)
2 Computer Science Department, Iqra University, Karachi 75300, Pakistan; [email protected] (H.F.);
[email protected] (N.I.)
3 Department Computer Science, Karachi Institute of Economics and Technology, Karachi 74600, Pakistan
4 Computer Science Department, Bahria University Karachi Campus, Karachi 75300, Pakistan;
[email protected]
5 Department of Network and Communication Engineering, College of Computer Science and Information
Systems, Najran University, Najran 61441, Saudi Arabia; [email protected]
6 Department of Information Systems, College of Computer Science and Information Systems,
Najran University, Najran 61441, Saudi Arabia; [email protected]
* Correspondence: [email protected]

Abstract: Forecasting rainfall is crucial to the well-being of individuals and is significant everywhere
in the world. It contributes to reducing the disastrous effects of floods on agriculture, human life, and
socioeconomic systems. This study discusses the challenges of effectively forecasting rainfall and
floods and the necessity of combining data with flood channel mathematical modelling to forecast
floodwater levels and velocities. This research focuses on leveraging historical meteorological data to
find trends using machine learning and deep learning approaches to estimate rainfall. The Bangladesh
Meteorological Department provided the data for the study, which also uses eight machine learning
algorithms. The performance of the machine learning models is examined using evaluation measures
like the R2 score, root mean squared error and validation loss. According to this research’s findings,
Citation: Rajab, A.; Farman, H.;
polynomial regression, random forest regression, and long short-term memory (LSTM) had the
Islam, N.; Syed, D.; Elmagzoub, M.A.;
highest performance levels. Random forest and polynomial regression have an R2 value of 0.76, while
Shaikh, A.; Akram, M.; Alrizq, M.
LSTM has a loss value of 0.09, respectively.
Flood Forecasting by Using Machine
Learning: A Study Leveraging
Historic Climatic Records of
Keywords: forecasting rainfall; RMSE; ANN; regression
Bangladesh. Water 2023, 15, 3970.
https://doi.org/10.3390/w15223970

Academic Editors: Marco Franchini

1. Introduction
and Gwo-Fong Lin
Natural calamities like hurricanes, earthquakes, floods, wildfires, and tsunamis are
Received: 23 September 2023 caused by the forces of nature and can happen suddenly. Environmental variables, such
Revised: 10 November 2023 as climate change, deforestation, and urbanisation, frequently feed these occurrences and
Accepted: 12 November 2023 increase their frequency and intensity. Natural catastrophes can have severe effects, leading
Published: 15 November 2023
to extensive destruction and fatalities. One of the most critical weather factors that affects
many parts of our everyday lives is rainfall [1,2]. Floods, one of the planet’s most common
catastrophes, seriously negatively impact the economy and agribusiness. They are regularly
Copyright: © 2023 by the authors.
observed when there is inadequate drainage and a lot of rain.
Licensee MDPI, Basel, Switzerland. Various kinds of rainfall exist, and unique mechanisms and climatic factors distinguish
This article is an open access article each. A few typical types of precipitation are mentioned in Figure 1. Water Supply [3],
distributed under the terms and Plant Growth and Agriculture, Erosion and Soil Moisture, Flooding [4–10], Water Quality
conditions of the Creative Commons and Pollution and Weather and Climate Patterns [11] are some notable effects of rainfall.
Attribution (CC BY) license (https:// The potential of using machine learning algorithms for flood/rainfall prediction and
creativecommons.org/licenses/by/ the importance of the problem cannot be denied. The severity and frequency of flood
4.0/). occurrences are expected to rise due to the dual threats of a fast-warming environment

Water 2023, 15, 3970. https://doi.org/10.3390/w15223970 https://www.mdpi.com/journal/water

 Water 2023, 15, 3970 2 of 37

and increasing urbanisation, endangering more people’s lives, ecological systems, and
economic systems. For many years, flood management strategies have been built on
the foundation of conventional flood forecasting techniques based on hydrological and
meteorological models and have produced significant insights. A more sophisticated,
adaptive, and data-driven methodology must be investigated, though, due to the complex
interaction of components that cause floods and the ever-increasing amount of data that
are accessible [12]. The field of artificial intelligence concentrates on creating machines
that can process data, learn from it, and make judgements. The use of machine learning
is an appealing approach for flood forecasting because it holds the promise of revealing
Water 2023, 15, x FOR PEER REVIEW 2 of 40
intricate, complicated correlations within huge datasets. Its capacity to incorporate data
from numerous sources, including satellite images, river gauge data, and climate models,
offers chances to improve floods’ precision, predictability, and lead time.

Figure 1. Various types of rainfall according to literature.

Figure 1. Various types of rainfall according to literature.
In this direction, a number of studies have been presented, such as those of Asif
etThe
al. [1], and Luo
potential et al. machine
of using [13]. Similarly,
learningeffective
algorithms machine learning has
for flood/rainfall been utilised
prediction and to
theconstruct
importance a rain forecast
of the problemmodel in numerous
cannot be denied. research [1,6,14].
The severity andOsmani et al.of[15]
frequency suggest
flood oc- a
novel approach
currences are expected for predicting
to rise duemonthly
to the dualdry threats
days at of
sixatarget points using
fast-warming several machine
environment and
learningurbanisation,
increasing (ML) techniques. Various other
endangering more studies,
people’ssuchlives,asecological
those of Manandhar
systems, and et eco-
al. [16],
Gude et al. [3], and Nguyen and Chen [4], have reported utilising
nomic systems. For many years, flood management strategies have been built on the foun- the concepts of fuzzy
logic,
dation ofsupport vectorflood
conventional and k-nearest
forecastingneighbours
techniquesapproach.
based onAswad et al. [10]
hydrological andproposed
meteoro-the
use of
logical an Internet-of-Things-driven
models and have produced significant flood status forecast
insights. A moreframework in order
sophisticated, to make it
adaptive,
andeasier to forecast
data-driven when riversmust
methodology would be flood. A similar
investigated, approach
though, duehasto been presented
the complex in Cihan
interac-
and Elif’s work [17].
tion of components that cause floods and the ever-increasing amount of data that are ac-
cessible Based
[12]. Theon field
a critical study of
of artificial the literature,
intelligence it has been
concentrates onobserved that workthat
creating machines on rainfall
can
process data, learn from it, and make judgements. The use of machine learning is anand
prediction is at infancy. Bangladesh is a region which is highly affected by rainfall ap-lots
of lives are lost yearly because of floods. The work on flood prediction
pealing approach for flood forecasting because it holds the promise of revealing intricate, for Bangladesh
has not been
complicated extensive.within
correlations This study performsIts
huge datasets. a comparative analysis of different
capacity to incorporate data fromML nu-and
merous sources, including satellite images, river gauge data, and climate models, offers for
deep learning (DL) methods for rainfall prediction. Then, we identify the best models
predicting
chances the rainfall
to improve floods’inprecision,
Bangladesh. Finally, this
predictability, paper
and leadoffers
time. a thorough investigation
ofInthe suggested model, for which a lengthy experiment
this direction, a number of studies have been presented, wassuchused. To summarise,
as those of Asif et al.the
significant contributions of the discussed work are as follows:
[1], and Luo et al. [13]. Similarly, effective machine learning has been utilised to construct
a rain Highlight
1. forecast modelthe serious and long-lasting
in numerous effects that
research [1,6,14]. Osmanifloodset have onsuggest
al. [15] the socioeconomic
a novel
approach for predicting monthly dry days at six target points using several machine learn- in
system, agriculture, and human life while acknowledging the growing challenge
ing (ML)accurately
techniques. estimating rainfall
Various other because
studies, of as
such climatic changes,
those of Manandharnon-linear
et al. qualities,
[16], Gudeand
variable attributes.
et al. [3], and Nguyen and Chen [4], have reported utilising the concepts of fuzzy logic,
2.
support Suggest
vector anda combined
k-nearestapproach
neighbours forapproach.
data with Aswad
computationally
et al. [10] intensive
proposedflood channel
the use of
mathematical modelsflood
an Internet-of-Things-driven to predict
statusflooding
forecast levels and velocities
framework in order across
to makea wide area.
it easier to
3.
forecast Identify undetected
when rivers would trends
flood. in
A historical meteorological
similar approach has been data to identify
presented in machine
Cihan and learn-
ing and
Elif’s work [17]. deep learning approaches as valuable tools for precisely estimating rainfall
with
Based onquantitative results
a critical study to prove
of the the usefulness
literature, it has beenofobserved
the machine thatlearning
work onmodels.
rainfall
4. Implement evaluation measures to assess the efficiency and
prediction is at infancy. Bangladesh is a region which is highly affected by rainfall progress made andby the
machine learning models, of such as the 2
R work
scoreonand rootprediction
mean squared error.
lots of lives are lost yearly because floods. The flood for Bangladesh
has not been extensive. This study performs a comparative analysis of different ML and
deep learning (DL) methods for rainfall prediction. Then, we identify the best models for
predicting the rainfall in Bangladesh. Finally, this paper offers a thorough investigation of
the suggested model, for which a lengthy experiment was used. To summarise, the signif-
icant contributions of the discussed work are as follows:
Water 2023, 15, 3970 3 of 37

5. Observation that the random forest regressor and k-nearest neighbours algorithms
achieved high accuracy, 96% and 99%, respectively.
The remaining sections of this work are organised as follows: Section 2 presents
a detailed literature review regarding the materials and methods, and partly examines
the theoretical background on flood early prediction utilising deep learning and various
machine learning techniques. Section 3 discusses the proposed methodology. Sections 4–6
present the results and discussion for comparing various machine learning and deep
learning techniques for flood earlier forecasting based on a dataset. Section 7 provides a
conclusion and future recommendations.

2. Literature Review
We structured this section as follows. First, the past studies are highlighted. A
discussion on the justification of this research follows this.

2.1. Related Work

Scholars have employed both qualitative and quantitative techniques to determine
flood exposure. One of the qualitative approaches utilised for identifying vulnerable
areas to flooding is the Analytic Hierarchy Process. Nevertheless, statistics-based and
machine-learning-related quantitative techniques can be divided into two groups. Over
the years, numerous mathematical and probabilistic algorithms have been employed for
predicting floods [18,19]. Various researchers [20,21] propose AI-based metaheuristics
algorithms to solve problems that have high complexity like security, load balancing,
resource optimisation and forecasting. The capability of ML-based approaches to manage
huge volumes of data has boosted interest in these algorithms for forecasting flooding over
the past few years. ML makes it possible to learn from historical data [22–25]. It also creates
models for forecasts based on historical data. The ability to predict floods will be assisted
by this method. We must first tell the system how to generate the output and outcomes.
But now, with the assistance of machine learning, it creates models and provides results
on its own. The majority of flood-related machine learning research either forecasts future
floods or aids in developing safety measures. Floods can be devastating in some years,
though, when there has been a lot of rain and water flowing upstream [26]. Kerala, an
Indian state in the south, saw a once-in-a-century flood. The cost of damage to both life
and property was considerable. This inspired us to conduct research on the rainfall pattern
in Kerala. Bangladesh has flooded every year, taking lives, livelihoods, crops, and property.
Flooding happens when a lake, river, or water overflows and engulfs neighbouring
land. Each year, floods affect over 4.84 million people in India, 3.84 million in Bangladesh,
and 3.28 million in China [27]. India is now one of the nations that has suffered the worst
floods, with the most recent calamity in Kerala in August 2018 being an exceptional in-
stance [11,28,29]. Over the years, much effort has been made to forecast the possibility of
flooding based on precipitation, humidity, temperature, water velocity, and other char-
acteristics using Internet of Things (IoT) and ML approaches. Nothing has attempted to
predict the likelihood of a flood depending on the temperature and severity of the rainfall,
which is the research’s main flaw. In contradiction to a model developed using machine
learning, the results show that a deep neural network may be employed effectively for
forecasting floods with the maximum accuracy based simply on monsoon characteristics
before flood occurrences. According to Adnan et al. [30], flood prediction has been a
significant area of study for scholars worldwide. This is because efficient and real-time
prediction of floods is essential for giving individuals who live close to flood zones the
warning they need to flee. Consequently, in this study, a 5 h flood forecast model for Kuala
Lumpur’s rainfall area was provided using an enhanced Neural Network Autoregressive
Model methodology. The 5 h NNARX flood water level forecast framework was created
using MATLAB Neural Network Toolkit. The findings showed that the NNARX model
effectively accurately estimated the flood water level five hours early. Mosavi et al. [31]
combine new ML techniques with traditional methods to create more precise and effective
Water 2023, 15, 3970 4 of 37

prediction models. This publication’s essential contribution is its discussion of the state
of ML models for flood prediction and recommendations for the most effective models to
use. To give a comprehensive picture of the many ML methodologies utilised in the area,
this study predominantly looks at the studies where ML models were assessed through
a qualitative review of reliability, accuracy, effectiveness, and performance. According to
Chen et al. [32], the area significance is divided into grids based on longitude and latitude,
and the data on precipitation and drainage collected at stations are combined into tensors
depending on station coordinates. Instead of a one-dimensional time sequence, the input
characteristic is a two-dimensional time series with spatial data.
According to Motta et al. [33], this effort will integrate machine learning classifiers with
Geographic Information Systems (GIS) methods to provide a flood prediction platform that
can be helpful for resiliency management. With the help of this approach, it is possible to
create realistic variables and risk indicators for the likelihood of flooding at the municipality
level, which may be used to create long-term plans for smart cities. According to a review
of past research articles, Maspo et al. [7] review the ML methods currently being utilised
for flood forecasting. This research tries to identify the most helpful flood forecast methods.
This research aims to list the critical variables and the most recent ML methods for flood
prediction. According to Sankaranarayanan et al. [13], the public and the government may
be able to plan both short- and long-term mitigation strategies, be ready for evacuation
and rescue operations, and provide relief for flood victims if they receive early warning
of a flood calamity. In this case, the location of the impacted areas and their respective
seriousness are two of the critical considerations in most flood mitigation methods. There
is still no reliable method for predicting floods in advance. Previous technologies usually
relied on prepared and manually entered data. Because the processes were time consuming,
making early and real-time projections was impossible. Innovative operational approaches
have been examined by Parag et al. [34]. The researchers observe and examine the current
trend regarding data-driven solutions for flood forecasting. For algorithms based on
machine learning that have been developed using historical data for climatic variables,
predicting jobs is growing more and more significant. A review offered by Furquim et al. [5]
presents the use of data gathered from urban rivers to forecast floods in an effort to decrease
the damage caused by floods. After the involvement hypothesis had shown the mutual
dependence of the information, the artificial neural networks were reviewed to see how
accurate their forecasting algorithms were. WSNs have been set up whenever serious
flooding-related issues have arisen. Adnan et al.’s [35] approach was suggested to develop
risk-based plans for growth and enhance current warning systems for emergencies. To
forecast and determine potential flooding locations or flood-sensitive regions in the Teesta
River basin, Talukdar et al. [6] applied cutting-edge novel ensemble machine learning
strategies. A rainfall forecasting algorithm employing the effective machine learning
random forest was proposed by Adnan et al. [30] by giving a flood prediction utilising a
range of machine learning techniques. Gauhar et al. [36] employed a variety of coefficients
of association for feature selection and the k-NN technique to forecast a flood. It is widely
established that quantifying and lowering the uncertainty associated with hydrologic
prediction is crucial for predicting the risk of flooding and making educated decisions [37].
The current paper thoroughly examines the Bayesian forecasting methods used in flood
forecasting. According to Haque et al. [38], 180 individual models were produced using
five different machine learning techniques based on multiple combinations of temporal
lags for input data and lead times in prediction. The 5772 km2 Someshwari-Kangsa sub-
watershed in Bangladesh’s North Central hydrological region was the subject of modelling.
Using conventional machine learning approaches, it is challenging to predict when it will
rain [39]. However, several studies have been presented that forecast rainfall using various
computer algorithms. Osmani et al.’s [15] innovative method for predicting monthly dry
days (MDD) at six target stations in Bangladesh makes use of a variety of ML techniques.
The datasets for monthly days without precipitation and monthly days with rainfall were
produced using a range of rainfall limitations. Manandhar et al. [16] recommend employing
Water 2023, 15, 3970 5 of 37

machine learning approaches to look into the long-term implications of flood prevention in
Bangladesh. Data from socioeconomic surveys and historical events (such as migration and
mortality) are available from 1983 to 2014. Yaseen et al.’s [40] emphasis is on the greater
necessity and duty of handling human-caused catastrophes. Humans invented an area of
science called artificial intelligence (AI) that could be applied in this situation.
Aakash et al. [41] thoroughly analyse and compare the many approaches and algo-
rithms that scholars have used to estimate precipitation. The core objective is to present
non-experts with access to the techniques and approaches used in rainfall forecasting. In
the present research, a flood vulnerability map for Iran is produced by utilising the concept
of the convolutional neural networks (CNN) method [42], one of the more current and
effective techniques in enormous datasets. In their discussion of several case studies, Sergey
et al. [43] use the example of ensemble-based storm surge simulation for forecasting floods
in St. Petersburg, Russia, to look at the opportunities presented by the established method-
ology. Mosavi et al.’s [31] main contribution is to show the current state of ML models for
flood prediction and to provide insight into the most appropriate models. India [11] has
some of the worst flood damage in the world right now, with the most recent disaster in
Kerala in August 2018 as a prime example. The problem is that no one has attempted to
forecast the likelihood of a flood using rainfall volume and temperature. Therefore, the
Neural Network has been utilised to forecast the likelihood of floods based on temperature
and rainfall intensity. Therefore, Gude et al. [3] propose flood prediction as one of the
primary topics to be researched in hydrology. Although many academics have studied
this problem using various approaches, such as physical models and image processing,
the accuracy and time steps still fall short for all applications. This study examines deep
learning techniques for gauging height and evaluates the associated uncertainty. Current
ML techniques for flood prediction are evaluated by Maspo et al. [7], and the parameters
utilised to predict floods are based on an analysis of past research publications. According
to Nevo et al. [44], the multidimensional model is a machine learning replacement for the
hydraulic modelling of flooding flows. Compared to past information, all models meet
expectations of performance that are high enough for use in operational situations. Mitra
et al. [8] offer an embedded system that utilises IoT-based machine learning to forecast
the possibility of floods in a river basin. The device uses a ZigBee connection to link the
WSN to a customisable mesh network, and then it uses a GPRS module to send data over
the internet.
Jeerana et al. [9] examine the possibility of using machine learning methods to fore-
cast flood occurrences in the Pattani River using open data. A probabilistic prediction
framework was created by Chen et al. [32] using several machine learning approaches.
Three techniques utilising multiple scenarios for decision-making were used along with
the ML based approaches to evaluate how well they were able to model the risk of flooding
in the Ningdu Catchment, which Khosravi et al. [45] address as one of China’s foremost
flood-prone geographic areas. El-Magd et al. [46] employed the extreme gradient boost and
KNN methods to produce flash flood prediction maps for the River El-Laquita in Egypt’s
eastern centre area. To predict floods on the shores of the rivers Daya and Bhargavi, which
flow across the Indian state of Odisha, Nayak et al. [47] use the Deep Belief Network (DBN).
In a comparative investigation, additional machine learning techniques are applied to fully
depict the effects of dam construction. Tayfur et al.’s [48] main concern is to present the
application of swarm-based optimisation, ant-colony optimiser, artificial neural network
(ANN) and genetic-algorithm-based approaches to flood hydrograph prediction.
To predict river flooding in the Barak River, Sahoo et al. [49] examined the corre-
sponding precision of radial basis functions. As reported by Qian et al. [50], there have
been considerable financial and human damages due to an increase in flash floods in
metropolitan areas. To precisely describe the specifics of flood improvement, the current
flood prediction techniques are either too sluggish or excessively straightforward. This
research uses deep neural networks to accelerate the mathematical calculation of a 2D
urban flood forecasting technique based on thermodynamics and controlled by the Shal-
Water 2023, 15, 3970 6 of 37

low Water Equation (SWE). Researchers retrieve flood patterns from data generated via a
partial differential equation (PDE) generator using convolutional neural networks (CNN)
and conditionally generative adversarial networks (cGANs). The four ML-based FSMs
mentioned by Adnan et al. [35] are random forest (RF), KNN, multilayer perceptron (MLP),
and hybridised genetic algorithm–gaussian radial basis function–support vector regression
(GA; RBF; SVR). Scott Miau and Wei-Hsi Hung [51] also utilise a deep learning framework.
Hossain et al. [52] describe an effort to create a system for analysing long-term seasonal
rainfall trends in Western Australia using Lavenberg–Marquardt multiple linear regression
and artificial-based methodologies. RBFs, including linear and nonlinear kernel parameters,
perform better in the same catchment under different circumstances. The response to lighter
precipitation would be very different from that to heavier one, which is a handy way to
reveal the dynamics of an SVM classifier. The study also shows an unexpected outcome in
the SVM response to diverse rainstorm-related inputs.
According to Aswad et al. [10], forecasting flood status is challenging and calls for in-
depth research into the underlying causes of floods. This study recommends a TpoT-based
model to help anticipate when rivers may flood. The IoT-FSP concept uses the Internet of
Things framework to facilitate flood data collection and three approaches for ML for flood
forecasts. As Ighile et al. [53] reported, this study forecasted the flood-prone locations in
Nigeria using historical flood records from 1985 to 2020 and plenty of conditioned variables.
An exact flood prediction model can be made using various machine learning techniques,
according to Kunvergi et al.’s [54] investigation. The generalised additive model (GAM),
the boosted regression tree (BTR), and the multivariate adaptive regression splines (MARS)
are three novel machine learning methods that Dodangeh et al. [55] suggested. These
models were built using random subsampling (RS), bootstrapping (BT), and multi-time
resampling techniques. The province of Ardabil, to which this approach was employed, is
situated near the Caspian Sea’s coast and frequently endures severe flooding.
The study by Khairudin et al. [56] aims to investigate the impact of various time-series
scales of rainfall information from eight rainfall stations along the Kelantan River on the
accuracy of the water level forecasts at Kuala Krai station. To create a flood forecasting
framework, Dtissibe, Francis Yongwa et al. [57] employed the multiple-layer perceptron
and flow as input–output parameters. For this, a set of data corresponding to the estimates
of rainfall recorded in Australia’s major cities throughout the previous ten years was
provided to the primary machine learning methods (kNN, decision tree, random forest,
and neural networks). Sarasa-Cabezuelo’s study [58] outlines a qualitative investigation
of using machine learning to predict the probability of rain. The outcomes demonstrate
that neural networks are the best approach. This work compares the efficacy of rainfall
forecasting techniques based on modern machine learning techniques for forecasting hourly
volumes of rainfall using weather time-series data from cities across the United Kingdom.
Liyew and Melese [59] analyse the performance of these algorithms. The analytical
hierarchical technique, a multi-parameter modelling tool, is used by MC Aydin and Bir-
inciolu [17] to analyse the risk of flooding assessments in the Turkish province of Bitlis.
Table 1 contrasts the pros and cons of various methods, datasets, and resources used in the
literature for rainfall or weather forecasting.

Table 1. A comparison of literature on rainfall prediction using machine learning.

Reference Dataset Model Pros Cons

A flood risk score was
Urban datasets from produced by combining the Hourly dataset; not a
Machine Learning
[33] January 2013 and results of the random forest consistent forecast for the
(RF) and GIS
December 2018 model and the Hot Spot entire year.
research.
Water 2023, 15, 3970 7 of 37

Table 1. Cont.

Reference Dataset Model Pros Cons

The outcomes demonstrate
Rainfall in April 2014 in Chaos theory MLP, Not a consistent forecast
[5] that the MLP outperforms the
Brazil E-RNN for the entire year.
ERNN.
A less extensive training
Yilan River basin and (SVR), fuzzy inference Statistical parameters are used dataset, a smaller feature
[4]
Taiwan 2012 to 2018 model (FIM), (k-NN) to analyse time series data. set, along with lower
precision and recall.
Dumdum weather Hybrid neural Selection of features: hybrid Less precision just for a
[25]
station framework neural system. tiny area.
Based on only one
India’s annual rainfall is Assist farmers in making the
characteristic, no
[29] included in the data Linear regression best decision for harvesting a
experimental results
collection. particular crop.
were discovered.
According to the results,
Used the dataset of 2016 random forest can make Small dataset utilised for
[14] DT, KNN, LR, NB, RF
to 2019 of Bangladesh. reliable forecasts for daily experiment.
rainfall estimates.
Multiple linear The created model was tested
Dataset reported of
regression (MLR) and extensively, and the results
[57] France. 2002 to 2018 Event-wise data analysis.
non-linear modelling demonstrated the usefulness of
events
technique, (ANN) forecasting.
Data on Indian summer Expert system based on Statistical parameters are used Less precision; covers a
[60]
monsoon rainfall fuzzy entropy to analyse time series data. much smaller area.
The Indian
Meteorological Institute Only 1- and 2-month
Artificial Neural
[61] in Pune collected data on Dataset with a long-time series. forward forecast;
Network (ANN)
North India’s monthly minimal feature set.
rainfall.

2.2. Discussion on Past Studies

Before proceeding further, let us discuss key findings obtained from the literature,
as follows:
• Classical meteorological systems: In the past, physical-based models like the HEC-
HMS, SWAT, and MIKE SHE predominated flood forecasting. These representations
incorporate physical equations that represent the flow and accumulation of water [2].
• Mathematical Models based on statistics: The exponential smoothing and ARIMA
statistical analysis of time series techniques were also employed to anticipate river
flows and flood levels. These techniques rely on the data’s statistical patterns [62].
• Incorporation of machine learning techniques: Machine learning techniques have
become more prevalent in recent years. Investigations have demonstrated that when
there are a lot of data available, machine learning can frequently match or even surpass
conventional hydrological projections [63].
Focusing on machine learning, LSTM and ANN are now favoured options for
flood forecasting.
Besides those discussed above, there have been several other developments and
breakthroughs when using machine learning (ML) for flood forecasting. However, there
are several gaps and restrictions in prior work [13,33,64,65]:
• Numerous studies rely on small or sparse datasets, which might not fully account
for all possible flood situations. Records for extreme occurrences or isolated flooding
disasters are frequently lacking. Climatic datasets are often incomplete and inaccurate,
particularly in nations with limited resources.
Water 2023, 15, 3970 8 of 37

• Several models are inefficient when used differently because they are overfit to datasets
or areas. Model portability between several geographic areas is still rugged.
• Numerous ML models, intense learning models, behave as “black boxes”, making it
challenging to comprehend how they make decisions. This makes it difficult to win
over the trust of stakeholders and end users.
• When ML models are combined with conventional meteorological models, which
have been widely used and relied upon for years, there is frequently a gap. The
physical processes that play a role in flood generation and propagation are not always
adequately taken into consideration in studies.
• For real-time prediction, the computational burden of some sophisticated ML models
may be too high. Specific models could be useless in time-sensitive situations due to
the time required for data collection, initial processing, and forecasting.
• Certain approaches might be practical in local watersheds or urban areas, but they
might have difficulties when scaled up to substantial river valleys.
Regardless of these challenges, there is a lot of scope for machine learning in flood
forecasting. To utilise machine learning’s maximum potential, it will be essential to carry
out ongoing research, collect data, engage stakeholders, and integrate ML with conventional
modelling techniques. This paper employs several machine learning and deep learning
models for rainfall prediction. We briefly discuss the rationale behind our choice.
For choosing ANN, the reason is that dynamic non-linear correlations in data, which
are frequently present in the hydrological system, can be captured using ANNs. A suffi-
ciently large ANN can hypothetically approximate any function. Because of this, they are
adaptable for various jobs, including flood forecasting. ANNs can change their architecture
(depth, width) to accommodate various datasets and prediction timeframes [57,66].
Similarly, the choice of LSMT is obvious. Flood forecasting is a time-series challenge
by nature. By preserving a “stored memory” of previous inputs in the internal neuron
states, LSTMs, a form of recurrent neural network (RNN), are created to handle sequential
information. The issue of vanishing gradients affects traditional RNNs, making it difficult
for them to learn dependence over time. LSTMs are less prone to this issue because of
their latching mechanisms, which enable them to learn and remember over lengthy peri-
ods [67]. Additionally, convolutional neural networks (CNNs) and other neural network
architectures can be integrated with LSTMs to capture spatial and temporal patterns [68,69].
The proposed study can forecast rainfall for every season and region in Bangladesh.
From Table 1, it is clear that most efforts are for various restricted regions and have some
significant flaws, like a relatively tiny dataset, a small feature set, and lower precision.
On the other hand, in this study, machine learning methods and a two-layer long short-
term memory (LSTM) method have been utilised [1,7,21] and an artificial neural network
(ANN) [33,58] for predicting rainfall in Bangladesh has been developed. It solves the
backflow problem found in other works. It uses a wider dataset with 18 features, the 16
most important of which are used. The proposed study can forecast rainfall in Bangladesh
for any season.

3. Proposed Methodology
This proposed research aims to determine whether, by utilising machine and deep
learning algorithms, a higher accuracy rate can be attained while also reducing error. The
dataset includes information on Bangladesh’s monthly and yearly rainfall (1949 to 2013)
index as well as information on the number of times a year that floods occur close to
35 stations: Khulna, Dinajpur, Bogra, Srimangal, Satkhira, Mymensingh, Jessor, Comilla,
Cox’sBaza, Faridpur, Barisal, Chittagong (IAP-Patenga), Maijdee, Court, Dhaka, Rangpu,
Sylhet, Rangamati, Ishurd, Rajshahi, Chandpur, Hatiya, Bhola, Sandwip, Patuakhali, Feni,
Khepupara, Madaripur, Kutubdia, Sitakunda, Teknaf, Tangail, Mongla, Chuadanga, Syed-
pur and Chittagong (City-Ambagan). Furthermore, the data were preprocessed using
feature engineering, data normalisation, and feature encoding. After splitting the dataset
into training and testing portions in an 80:20 ratio, applying the machine learning model
 edpur and Chittagong (City-Ambagan). Furthermore, the data were preprocessed
feature engineering, data normalisation, and feature encoding. After splitting the d
into training and testing portions in an 80:20 ratio, applying the machine learning m
Water 2023, 15, 3970 9 of 37
is essential. For comparison, it is necessary to use models like the k-nearest neigh
support vector machine, decision tree regressor, random forest mode, AdaBoostR
sor, Stacking Regressor,
is essential.and artificial itneural
For comparison, network.
is necessary Finally,
to use models the model
like the k-nearest that is b
neighbour,
support vector machine, decision tree regressor, random forest mode, AdaBoostRegressor,
predicting floods can beRegressor,
Stacking identified basedneural
and artificial on network.
the RMSE Finally,and R2 scores
the model ofatthe
that is best models
predict-
2
Figure 2 shows theingworkflow
floods can be identified based on the RMSE and
of the methodology in Rdetail.
scores of the models used. Figure 2
shows the workflow of the methodology in detail.

Figure 2. Machine learning pipeline

Figure 2. Machine describing
learning the proposed
pipeline describing methodology.
the proposed methodology.

Figure 3 represents the complete architecture of the proposed work. The only goal of
this research is to use deep learning and supervised learning to achieve maximum accuracy.
The points in the validation set are used to determine the accuracy of the regressor following
learning with the training data.
 Water 2023, 15, x FOR PEER REVIEW 10 of 40

Figure 3 represents the complete architecture of the proposed work. The only goal of
Water 2023, 15, 3970 this research is to use deep learning and supervised learning to achieve maximum accu- 10 of 37
racy. The points in the validation set are used to determine the accuracy of the regressor
following learning with the training data.

Figure 3. Block diagram describing the proposed system.

Figure 3. Block diagram describing the proposed system.

3.1.3.1. DatasetDescription
Dataset Description
The data are acquired from Bangladesh’s Weather Department in Dhaka, the main
The data are acquired from Bangladesh’s Weather Department in Dhaka, the main
authority for tracking and making predictions for every natural catastrophe to reduce
authority for tracking and making predictions for every natural catastrophe to reduce
mortality. To deliver accurate forecasts for the weather, we want to learn what we can
mortality. To deliver accurate forecasts for the weather, we want to learn what we can
deduce about previous times and how it connects to present-day climate change and the
deduce
general trendsprevious
about times
of the planet’s and how
weather it connects
by using to present-day
this dataset. climate
From 1948 to 2013, change and
the dataset
theincludes
generalcomprehensive
trends of themonthly
planet’saverages
weatherforbyBangladesh
using thisthat
dataset.
are area-specific forto
From 1948 2013, the
max-
dataset includes comprehensive monthly averages for Bangladesh that
imum temperature, minimum temperature, rainfall, relative humidity, wind speed, cloud are area-specific
forcover,
maximum temperature,
and brilliant minimum
sunshine. temperature,
Also included rainfall, station
are the weather relativenumbers,
humidity, windY speed,
X and
cloud cover, and
coordinates, brilliant
latitude, sunshine.
longitude, andAlso included
altitude. are the weather
This research develops station numbers,
and evaluates the X and
top eight deeplatitude,
Y coordinates, learninglongitude,
and machine andlearning models
altitude. Thisusing our dataset.
research developsFigure
and 4evaluates
shows the
topthe snapshot
eight deepof the dataset.
learning andThe data forlearning
machine Bangladesh’s
modelsmaximum
using ourand dataset.
minimumFigure
monthly 4 shows
thetemperatures
snapshot ofand the annual
dataset.rainfall are shown
The data in Figures 5 and
for Bangladesh’s 6, respectively,
maximum throughout
and minimum monthly
the entire 65-year period. As can be seen, the maximum temperature in Bangladesh
temperatures and annual rainfall are shown in Figures 5 and 6, respectively, throughout is in
Water 2023, 15, x FOR PEER REVIEW April, March or May, whereas the lowest temperature occurs in January or December. 11 of 40
the entire 65-year period. As can be seen, the maximum temperature in Bangladesh is in
April, March or May, whereas the lowest temperature occurs in January or December.

Figure 4. Snapshot of the dataset [70]

Figure 4. Snapshot of the dataset [70].
Water 2023, 15, 3970 11 of 37

Figure 4. Snapshot of the dataset [70]

(a)

(b)
Figure
Figure5.5.Yearly
Yearlyand
andmonthly
monthlyrainfall
rainfallininBangladesh:
Bangladesh:(a)(a)yearly
yearlyrainfall
rainfallininBangladesh,
Bangladesh,(b)
(b)monthly
monthly
rainfall in Bangladesh.
rainfall in Bangladesh.

These figures show the bar plot, monthly precipitation, and the time series of weather
forecasting. According to the figure, the middle months experience the highest frequency
of rainfall during a 12-month period, which steadily declines afterwards. Projecting future
values using past data is known as time series forecasting. The graphs indicate that the
peak of rainfall progressively rises after a few years. The highest temperature in Bangladesh
is depicted in Figure 6. The bar plot shows the variation in the peak temperature over
time. The varying patterns of the bar indicate periods when the maximum temperature
was abnormally high or low. The right graph displays the variations in the minimum
temperature over time. The variation in the count plot indicates the seasonal variations in
the minimum temperature.
WaterWater 15, x15,
2023,2023, FOR3970
PEER REVIEW 12 of 12
40of 37

(a)

(b)
Figure 6. Bar
Figure diagram
6. Bar for for
diagram monthly minimum
monthly and
minimum maximum
and temperature
maximum temperaturein
in Bangladesh: (a)maxi-
Bangladesh: (a)
maximum temperature,(b)
mum temperature, (b)minimum
minimumtemperature.
temperature.

These figures show

Designing the bar plot,and
a deep-learning- monthly precipitation, and the
machine-learning-based time series
strategy of weather
for predicting rain is
the goal According
forecasting. of this endeavour. To laythe
to the figure, themiddle
foundation
monthsforexperience
building this themodel,
highestthe Bangladesh
frequency
Meteorological
of rainfall Department’s
during a 12-month (BMD)
period, whichdataset with
steadily 21,120 afterwards.
declines records wasProjecting
employed.future
Figure
values using 7’s data
past bar plot shows as
is known that theseries
time city inforecasting.
BangladeshThe withgraphs
the highest rainfall
indicate thatamount
the
is Teknaf, followed by Sylhet. Teknaf is a city located in the southeastern
peak of rainfall progressively rises after a few years. The highest temperature in Bangla- part of Bangladesh.
deshThe heavy rainfall
is depicted in Teknaf
in Figure 6. Theisbar
caused by various
plot shows things, including
the variation in the peaktopography,
temperature geographic
over
features,
time. and orographic
The varying patterns ofeffects.
the barThese elements
indicate work
periods together
when to make Teknaf
the maximum experience
temperature
wasmore rainfall than
abnormally high other
or low.areas
Theofright
Bangladesh. Additionally,
graph displays there is the
the variations inpotential for yearly
the minimum
 15 LATITUDE float nate of the sta-
tions.
tion.

longitude
The longitude of rainfall in specific locations and weather condi-
16 LONGITUDE float coordinate of the
tions.
Water 2023, 15, 3970 station. 13 of 37

This column likely represents the altitude or elevation of each

17 ‘ALT’, num metres
weather station.
variation in weather conditions and rainfall amounts, which a variety of weather-related
Rainfall measurements
variables are gathered
and climate change mayor recorded
impact. at a specific
Details of the eighteen
18 ‘Period’ floatfeatures are provided in
numeric
time step or period
Table 2 below. (year and month combined).

Figure 7. Bar plot for rainfall in Bangladesh.

Figure 7. Bar plot for rainfall in Bangladesh.

To perform automated regression evaluations, this work uses machine learning and
Table 2. Features used for training machine learning models.
deep learning techniques. The information from the meteorological department is used to
Sr. No. Attribute
forecast rainfall inAttributes
Bangladesh. The collection includes several
Description Type
characteristics. However,
Measurement
the output class for the forecast uses an attribute called “Rainfall”. The data collection’s
1. ‘Unnamed: 0’, This is likely an index or identifier column for the dataset. integer serial no
overall histogram illustration is shown in Figure 8. Following the data collection, several
2. ‘Station Names’ The name of the city or station where the flood occurred string categorical
preprocessing procedures
This column represents are for
the year carried
whichout, including
the weather verifying values, handling, scaling,
data
3. ‘Year’, and integer numerical
are transforming
recorded. some characteristics, like station names, etc. This will permit the super-
4. ‘Month’, vised learning regression algorithms
The month of the recorded data. to provide more accurate predictions.
integer Figure 8 rep-
numerical
5. ‘Max Temp 0C’ The maximum recorded temperature of a day.
resents the bar plot for rainfall in Bangladesh’s cities. float degrees Celsius
The minimum temperature experienced on a specific day
6. ‘Min Temp 0C’ float degrees Celsius
(degrees Celsius).
7. ‘Rainfall (mm)’, The amount of rainfall recorded (millimetres). float millimetres
This column represents the relative humidity recorded for a
8. relative humidity float percentage
specific month and year.
Wind speed in a particular direction at a given location and
9. ‘Wind Speed’, float metres per second.
time
This column represents the cloud coverage or cloudiness level
10 Cloud Coverage float percentage
recorded for a specific month and year.
This column contains the duration of bright sunshine recorded
11 ‘Bright Sunshine’ for a specific month and year. It represents when the sun is float measured in hours
visible, or the sky is clear.
In meteorology and weather monitoring, a station number is a
12 ‘Station Number’ unique identification declared to an individual weather station integer numerical identifier
or monitoring place.
Water 2023, 15, 3970 14 of 37

Table 2. Cont.

Sr. No. Attribute Attributes Description Type Measurement

This column could represent the X-coordinate or longitude coordinates of the
13 ‘X_COR’, float
values associated with the location of each weather station. station
This column could represent the Y-coordinate or latitude coordinates of the
14 Y_COR float
values associated with the location of each weather station. station
The latitude of rainfall on specific locations and weather latitude coordinate
15 LATITUDE float
conditions. of the station.
longitude
The longitude of rainfall in specific locations and weather
16 LONGITUDE float coordinate of the
conditions.
station.
This column likely represents the altitude or elevation of each
17 ‘ALT’, num metres
weather station.
Rainfall measurements are gathered or recorded at a specific
18 ‘Period’ float numeric
time step or period (year and month combined).

To perform automated regression evaluations, this work uses machine learning and
deep learning techniques. The information from the meteorological department is used to
forecast rainfall in Bangladesh. The collection includes several characteristics. However,
the output class for the forecast uses an attribute called “Rainfall”. The data collection’s
overall histogram illustration is shown in Figure 8. Following the data collection, several
preprocessing procedures are carried out, including verifying values, handling, scaling, and
transforming some characteristics, like station names, etc. This will permit the supervised
learning regression algorithms to provide more accurate predictions. Figure158 represents
Water 2023, 15, x FOR PEER REVIEW of 40
the bar plot for rainfall in Bangladesh’s cities.

Figure 8. Histogram depicting the feature values of the dataset.

Figure 8. Histogram depicting the feature values of the dataset.
3.2. Dataset Preprocessing or Cleaning
A preliminary processing data mining technique turns unstructured, incorrect input
into a format that the model can use and understand. Raw data are uneven, lacking many
key aspects, and full of errors. According to data exploration and estimation, there are no
redundant, invalid, or null values in the raw data for the used model. It is necessary to
Water 2023, 15, 3970 15 of 37

3.2. Dataset Preprocessing or Cleaning

A preliminary processing data mining technique turns unstructured, incorrect input
into a format that the model can use and understand. Raw data are uneven, lacking many
key aspects, and full of errors. According to data exploration and estimation, there are no
redundant, invalid, or null values in the raw data for the used model. It is necessary to
choose only the features pertinent to our model for forecasting rainfall during the prepro-
cessing phase of feature selection. This reduces training time and raises the accuracy of the
model. Table 3 and Figure 9 show the correlation coefficients of feature rainfall with various
variables. Then, the work employs the dropping of features. Correlation is calculated
between dependent and independent variables that are further used for modelling. The
Water 2023, 15, x FOR PEER REVIEWfollowing columns were dropped. These features have the lowest correlation with 17 ofthe
40
rainfall variable, as seen in Table 3.

Figure 9. Pictorial representation of correlation between features.

Figure 9. Pictorial representation of correlation between features.
3.6.
• Machine Learning
YEAR: This Modelsthe year;
represents
• One of the
X_COR: Thismost popularthe
represents and effective types
x-coordinate whereofrainfall
algorithmic learning is supervised
is happening;
machine
• learning,
Y_COR: and the types
This represents of machine learning algorithms are presented in Figure
the y-coordinate;
10.
• When we haveThe
LATITUDE: a few instances
latitude of location;
of the features–label pairings and want to predict a specific
outcome
• or label from
LONGITUDE: Thealongitude
set of features, we apply supervised learning. Our training set,
of the location;
•
which consists
ALT: of these of
The altitude features–label
the location.pairings, is used to create a machine learning model.
Our objective is to accurately predict new, unforeseen data. The most effective machine
learning and deep learning methods for analysing the daily rainfall quantity forecasts
have been chosen after evaluating many articles on rainfall prediction [8–20].
Water 2023, 15, 3970 16 of 37

Table 3. Correlation coefficients of rainfall with various variables.

Max Min Cloud Bright

Wind Station X_ Y_ LATI- LONGI-
Unnam YEAR Month Temp Temp Rainfall Humidity Cover- Sun- ALT Period
◦C ◦C Speed Num COR COR TUDE TUDE
age shine
0.064153 0.025109 0.132680 0.256821 0.596625 1.000000 1.0000 0.316366 0.766821 −0.673333 0.113804 0.167625 −0.066154 −0.105569 0.197805 −0.009696 0.02536
Water 2023, 15, 3970 17 of 37

3.3. Data Normalisation

Data normalisation is scaling the values of variables so that they have the same
quantitative weight and lie in an identical interval or scale. They rescale the entire dataset
to a standard range or distribution.
The equation for data normalisation using the min–max scaling technique is as follows:

x − min( x )
xnormalized = (1)
max( x ) − min( x )

where x represents the original value in the dataset, min(x) is the minimum value of the
dataset, max(x) is the maximum value of the dataset and, xnormalized is the normalised value
of x within the range [0, 1]. Equation (1) scales the values of the entire dataset to the range
[0, 1],

3.4. Feature Encoding

Feature encoding is an essential stage in data preprocessing, specifically for machine
learning and deep learning. It entails converting textual or category information into a
numerical representation that algorithms can utilise. The dataset’s primary feature, “Station
Names”, which provides string-type data, needs attention. Although machine learning
algorithms perform better with numbers, textual data are encoded. The attribute “Station”
contains a list of all the station names from which the daily rainfall statistics have been
gathered. Without adding any further fields, the category values of the “Station” variable
were then translated using label encoding into numerical values. There are numerous
varieties of feature-encoding methods. In our experiment, we employ label encoding,
which gives each category a distinct integer and is generally applied to data with ordinal
values. Label encoding assigns a unique integer to each station name.

3.5. Feature Scaling

Feature scaling is the scaling of individual features to have similar magnitudes within
the dataset. To ensure that the dataset was fair and applicable to the models utilised, the
dataset’s features were scaled using the Standard Scaler. Equation (2) for feature scaling
using the z-score scaling or standardisation technique is as follows, where feature scaling is
independently performed on individual features:

x−µ
xscaled = (2)
σ
where x = original value of the feature, µ = mean (average) of the feature in the dataset,
σ = standard deviation of the feature in the dataset, and xscaled = scaled value of the feature.

3.6. Machine Learning Models

One of the most popular and effective types of algorithmic learning is supervised
machine learning, and the types of machine learning algorithms are presented in Figure 10.
When we have a few instances of features–label pairings and want to predict a specific
outcome or label from a set of features, we apply supervised learning. Our training set,
which consists of these features–label pairings, is used to create a machine learning model.
Our objective is to accurately predict new, unforeseen data. The most effective machine
learning and deep learning methods for analysing the daily rainfall quantity forecasts have
been chosen after evaluating many articles on rainfall prediction [8–20].
The used dataset falls under the category of a regression problem since rainfall pre-
diction is a continuous number, or what programmers refer to as a floating-point number.
In this study, the regression algorithm is used to train the dataset. Eight machine learning
algorithms (polynomial linear regression, multiple linear regression, k nearest neighbours’
regression, regression with decision trees, support vector machine, random forest model,
Ada boost regression, and stacking regression) were tested and compared using real-time
 3.6. Machine Learning Models
One of the most popular and effective types of algorithmic learning is supervised
machine learning, and the types of machine learning algorithms are presented in Figure
10. When we have a few instances of features–label pairings and want to predict a specific
Water 2023, 15, 3970
outcome or label from a set of features, we apply supervised learning. Our training set,
18 of 37
which consists of these features–label pairings, is used to create a machine learning model.
Our objective is to accurately predict new, unforeseen data. The most effective machine
learning and deep learning methods for analysing the daily rainfall quantity forecasts
environmental data
have been chosen to forecast
after the many
evaluating daily intensity ofrainfall
articles on the rainfall. The algorithms
prediction [8–20]. with the
best accuracy were reported.

Figure 10. Various

Figure 10. Various types
types of
of machine
machine learning
learning algorithms
algorithms proposed
proposed in
in the
the literature.
literature.
3.6.1. Polynomial Linear Regression
A variation on simple linear regression, polynomial linear regression, enables more
intricate connections between the predictor variables (features) and the variable of interest.
The predictors are changed by being raised to different powers to create polynomial terms
in polynomial linear regression. Equation (3) represents polynomial linear regression, as
follows:
y = β0 + β 1 x 1 + β 2 x 2 + · · · + β n x n + e (3)
where y represents the target variable (dependent variable), x denotes the predictor variable
(independent variable), β 1 . . . β n are coefficients to be estimated, and e is the error term by
introducing higher-order terms (x1 ,x2 , x3 . . .x n ). A polynomial linear regression can capture
nonlinear relationships between the predictors and the target variable.

3.6.2. Multiple Linear Regression

With many predictor variables, multiple linear regression attempts to forecast the tar-
get variable. It requires that the predictors and the target variable have a linear relationship.
Multiple linear regression’s equations may be written as in Equation (4):

y = β 0 + β 1 x1 + β 2 x2 + · · · + β n x n + e (4)

where y represents the target variable (dependent variable), x1 and x2 denotes the predictor
variable (independent variable), β 0 , β 2 . . .β n are coefficient to be estimated, and e is the
error tem. Figure 11 shows the results using multiple linear regression actual and predicted
rainfall minimum temperature in terms of R2 and RMSE score.

3.6.3. K-Nearest Neighbours Regressor

In order to deal with classification and regression forecasting problems, the KNN
method is used. It falls within the category of supervised machine learning. The KNN
method uses feature similarity to predict the values of new data points, which implies
that the value of the new data point will depend on how closely it resembles the points
in the training set. Although there are numerous distance functions, Euclidean is the
most widely used one for calculating distance. The given distance equations for two- and
multidimensional values are shown in Equations (5) and (6), as follows:
 where y represents the target variable (dependent variable), x denotes the predictor vari-
able (independent variable), 𝛽 …𝛽 are coefficients to be estimated, and 𝜖 is the error
term by introducing higher-order terms (𝑥 ,𝑥 , 𝑥 …𝑥 ). A polynomial linear regression
can capture nonlinear relationships between the predictors and the target variable.
Water 2023, 15, 3970 19 of 37
3.6.2. Multiple Linear Regression
With many predictor variables, multiple linear regression attempts to forecast the
targetFor
variable. It requires that the predictors and the target variable have a linear rela-
two dimensions:
tionship. Multiple linear regression’s
rh equations may be writtenii as in Equation (4):
2 2
d= ( x − x ) + (y − y )
𝑦 = 𝛽 + 𝛽 2𝑥 +1 𝛽 𝑥 + 2⋯ + 1𝛽 𝑥 + 𝜖
(5)
(4)
where y represents
For the target variable (dependent variable), 𝑥 and 𝑥 denotes the predic-
multiple dimensions:
tor variable (independent variable), 𝛽 ,𝛽 …𝛽 are coefficient to be estimated, and 𝜖 is
r
the error tem. Figure 11 shows the results using
m
h multiplei linear regression actual and pre-
d = ∑
dicted rainfall minimum temperature in terms(of
i = 1
xi −
R2yand
2
i ) RMSE score. (6)

Water 2023, 15, x FOR PEER REVIEW 19 of 40

Figure 11. Results illustrating R2 and RMSE values for various machine learning algorithms.

Figure 11. Results illustrating R2 and RMSE values for various machine learning algorithms.

3.6.3. K-Nearest Neighbours Regressor

In order to deal with classification and regression forecasting problems, the KNN
method is used. It falls within the category of supervised machine learning. The KNN
method uses feature similarity to predict the values of new data points, which implies that
Water 2023, 15, 3970 20 of 37

3.6.4. Decision Tree Regressor

A supervised learning approach for regression tasks is called the Decision Tree Regres-
sor. It is a non-parametric approach that creates a model that resembles a tree in order to
generate predictions using a set of decision rules deduced from training data.

3.6.5. Support Vector Machine

A supervised learning approach extends the core concept of the Support Vector Ma-
chine (SVM) method to accommodate continuous target variables. To convert the provided
data characteristics into a higher dimension space, SVR employs a core implementation
of kernel. The decision boundary’s form and the model’s flexibility are influenced by the
selection of the main function. In our research, we use radial basis function for execution.

3.6.6. Random Forest Model

An adaptable, simple machine learning method called random forest typically pro-
duces outstanding results even without the use of hyper-parameter modification. It has
become one of the most extensively used methods due to its simplicity and adaptability (it
can be used for regression and classification).

3.6.7. AdaBoostRegressor
A supervised machine learning approach called AdaBoostRegressor is used for regres-
sion problems. It is an adaptation of the AdaBoost (Adaptive Boosting) technique, which
combines a number of weak learners (regression models) to produce an ensemble model
that is robust.

3.6.8. Stacking Regressor

The Stacking Regressor is a method based on machine learning that mixes various
separate regressor algorithms to build a meta-regressor for regression problems. It is based
on the idea of “model stacking”, in which inputs from several models are combined to
create the final prediction using an advanced model.

3.7. Deep Learning Model

Relying on this kind of training data and the learning goals, deep learning models
can be divided into supervised and unsupervised learning, as shown in Figure 12. We
used in our research, for the prediction of rainfall, supervised deep learning. Supervised
deep learning, input features and their matching target labels are supplied, and the models
are trained using labelled data. The objective is to learn the relationships between the
provided features and the desired output labels. Several well-liked supervised deep
learning architectures are as follows: artificial neural network (ANN), convolutional neural
network (CNN) and recurrent neural network (RNN). Multiple-layered neural networks
are used in deep learning algorithms and machine learning approaches, which enable the
model to automatically extract representations and patterns from data. Supervised and
unsupervised learning are the two main types of deep learning. A labelled dataset is used
to train the model in supervised learning, where each data point is linked to a specific
target or label. Training a model on an unlabelled dataset—one without predetermined
goal labels—is known as unsupervised learning. The dataset in this study has been
trained using the deep learning technique. Using real-time environmental data, three deep
learning algorithms—recurrent neural network (RNN), long short-term memory (LSTM),
and artificial neural network (ANN)—were evaluated and contrasted in order to predict
the typical intensity of the rainfall. The algorithms with the best accuracy were reported.
Water 2023, 15, 3970 21 of 37
Water 2023, 15, x FOR PEER REVIEW 21 of 40

Water 2023, 15, x FOR PEER REVIEW 21 of 40

Figure
Figure12.12.
Various types
Various eep
of dof
types learning
deep algorithms.
learning algorithms.

3.7.1.
3.7.1.Artificial
ArtificialNeural Network
Neural (ANN)
Network (ANN)
Figure 12. Various types
AAcomputational
computational of
model deep
model learning
called
called analgorithms.
an artificial neural
artificial network
neural (ANN)
network is motivated
(ANN) is motivatedby by the
the organisation and operation of the neural networks in the human brain.
organisation and operation of the neural networks in the human brain. It is a particular kindIt is a particular
3.7.1.
kind ofArtificial
ML approachNeural Network (ANN)
of ML approach thatthat
cancan consider
consider input
input data
data while
while making
making predictionsororchoices.
predictions choices. The input
The inputA layer, hidden layer(s),
computational model and output
called an layer areneural
artificial the three primary
network types of
(ANN) is layers in
motivated by (as
layer, hidden layer(s), and output layer are the three primary types of layers in an ANN
an ANN
the (as shownand
organisation in Figure 13). of
operation The weights
the neural indicate
networks theinconnections
the human between
brain. It neurons
is a particular
shown in Figure 13). The weights indicate the connections between neurons and specify
and
kindspecify
of ML theapproach
strength or significance
that of theinput
canofconsider information transmitted
datatransmitted
while making between them. or
predictions Dur-
choices.
the
ing strength
the training or significance
phase, the ANN the information
updates the weights of the connections between
based on them.
the During the
given
The input
training layer,
phase, hidden
the ANN layer(s),
updates andtheoutput
weights layer
of are
the the three
connections primary
based types
on of
the layers
given in
input
input data and the needed output. The error between the projected and actual output is
an
data ANNand (as
theshown
needed in Figure
output. 13).
The The weights
error betweenindicate
the the connections
projected and between
actual neurons
output is often
often transmitted backwards across the network to update the weights in a procedure
and specify the strength or significance of the information transmitted between them. Dur-
transmitted
known backwards across
as backpropagation. The ANN thelearns
network to update
and develops itsthe weights
prediction or in a procedure known
decision-mak-
ing the training phase, the ANN updates the weights of the connections based on the given
as skills
ing backpropagation. The ANN
thanks to this iterative learns
approach and develops
[8,57,61]. The net inputits prediction
for the general or artificial
decision-making
input data andmodel
neural
the needed output. The errorcomputed
between by theusing
projected and (7) actual output is
skills network
thanks to thismentioned below can be
iterative approach [8,57,61]. The Equations
net input for the and (8), artificial
general
asoften transmitted
follows: backwards across the network to update the weights in a procedure
neural network model mentioned below can be computed by using Equations (7) and (8),
known as backpropagation. The ANN learns and develops its prediction or decision-mak-
as follows:
ing skills thanks to this iterative approach [8,57,61]. The net input for the general artificial
neural network model mentioned below can be computed by using Equations m (7) and (8),
as follows: Yin = x 1 · w 1 + x 2 · w 2 + x 3 · w 3 + . . . . . . + x m · w m i.e., Yin = ∑ x i · wi (7)
i

Figure 13. Artificial neural network.

Figure Artificialneural
Figure 13. Artificial neuralnetwork.
network.
Water 2023, 15, 3970 22 of 37

Applying the activation function to the net input allows for the output estimation.

Y = F (Yin ) (8)

3.7.2. Recurrent Neural Network (RNN)

The RNN [5] analyses time series or sequential data. Instead of processing input
data in a single pass from input to output like feedforward neural networks do, RNNs
have a feedback loop that enables information to persist over time and be shared between
temporal stages (Figure 14). An RNN’s internal capacity to maintain a state or memory
enables it to recognise dependencies and patterns in sequential input. Through every time
step, its internal state is changed while considering both the recent and the last input state.
It can use data from earlier time steps to make predictions or generate outputs influenced
by the entire input sequence. A recurrent neural network (RNN) has various parameters
that govern the network’s behaviour and properties. The critical RNN parameters are
as follows: Input size: The dimensionality of input. It determines the number of RNN
inputs. The hidden size determines the number of memory cells in the RNN. It indicates
the dimensionality of the network’s internal state or memory. The output size specifies the
dimensionality of the output at each time step. It determines the number of RNN output
nodes—Sigmoid, tanh, and ReLU are examples of standard activation functions. Weight
matrix: Weight matrices manage the flow of information between the input, hidden, and
output layers in RNNs. These matrices are learnt throughout the training process and
regulate how each layer affects the others. The initial concealed state is the beginning
point for the internal state of the RNN. It can be set to zero or learnt during training as a
parameter. Recurrent weight matrix: The recurrent weight matrix relates the last time step
hidden state to the current time step hidden state. It governs how the memory of the RNN
is refreshed and transmitted over time. Bias Terms: At each layer, bias terms are added to
the weighted inputs to induce an offset or bias in the computation. They enable the RNN
to learn various intercepts for various characteristics. The following equations regulate the
computations of an RNN. For hidden state computation, Equation (9) is used:

h(t) = f (Wxh ∗ x (t) + Whh ∗ h(t − 1) + bh ) (9)

Output computation is performed by using Equation (10):

 
y(t) = f Why ∗ h(t) + by (10)

where t = time step, x(t) = input at ‘t’, and h(t) = hidden state at ‘t’. Wxh , Whh , Why are
weight matrices that control the flow of information, and bh and by are bias vectors.

3.7.3. Long Short-Term Memory

This framework is used to prevent the issue of vanishing gradients and preserve
persistent dependencies in sequential provided data. The input is represented by x(t) at
time t. The hidden state from the previous time step (t − 1) is represented by h(t − 1). Cell
state: the cell state from the prior time step (t − 1) is represented by c(t − 1). Output: at
time t, the hidden state or output is represented by h(t). The hidden state in an LSTM (Long
Short-Term Memory) represents the output or information propagated to the next time step
in the sequence (as shown in Figure 14). It serves as a memory of the previous time steps
and captures relevant information from the input sequence. The work for an LSTM unit is
as follows: for the forget gate, we use Equation (11):
 
f (t) = sigmoid W f ∗ [h(t − 1), x (t)] + b f (11)
2023, 15, x FOR PEER REVIEW 23 of 4
Water 2023, 15, 3970 23 of 37

Figure 14. Illustration of actual and predicted rainfall using multiple linear regression.

For the input gate, we use Equations (12)–(16):

i (t) = sigmoid(Wi ∗ [h(t − 1), x (t)] + bi ) (12)

Figure 14. Illustration of actual and predicted rainfall using multiple linear regression.
Ĉ(t) = tanh(Wc ∗ [h(t − 1), x (t)] + bc ) (13)
3.7.3. Long Short-Term Memory
This framework is used to prevent the issue of vanishing gradients and preserve per
sistent dependencies in sequential provided data. The input is represented by x(t) at tim
Water 2023, 15, 3970 24 of 37

Updating cell state works as follows:

c(t) = f (t) ∗ c(t − 1) + i (t) ∗ Ĉ(t) (14)

Output gate calculations are as follows:

o (t) = sigmoid(Wo ∗ [h(t − 1), x (t)] + bo ) (15)

Hidden state calculation is carried out as follows:

h(t) = o (t) ∗ tanh(c(t)) (16)

3.8. Implementation Details

We can efficiently capture the correlations between meteorological factors and rainfall
by utilising machine learning algorithms like regression models, decision trees, random
forests, or support vector machines. We can estimate rainfall with a respectable degree
of accuracy by taking advantage of these models’ strengths to learn from previous data
and make predictions based on input features. Convolutional neural networks (CNNs),
recurrent neural networks (RNNs), and long short-term memory (LSTM) networks are
examples of deep learning techniques that are particularly good at detecting complex
connections and patterns in sequential or spatial data. These models are ideally suited
for applications requiring rainfall prediction because they can easily handle the temporal
or spatial character of rainfall patterns. Deep learning algorithms may learn complicated
representations and produce precise forecasts by training on previous rainfall data. In our
research work, we implemented both techniques for producing a better rainfall model. The
steps for implementation are as follows. Table 4 summarises the details.

Table 4. Details of implementation.

Parameter Values
Framework Sk-learn, tensorflow
Training, validation, testing 60%, 20%, 20%
Number of epochs 30
Stopping criterion Early stopping
Activation functions ReLu
Optimiser Adam
Validation criterion 3-fold cross validation

• The various machine learning models used in this study have been implemented using
sk-learn.
• For deep learning models, tensorflow has been used.
• For optimising and finetuning of various hyperparameters, k-fold cross validation has
been performed.
• In addition, earlystopping callback of tensorflow has been used.
• The models are trained for 30 epochs.
• The validation split is 20%.
• The ReLu activation function has been used along with Adam optimiser.
As mentioned in the table, a resampling method called 3-fold cross-validation is used
to assess machine learning models on a small sample of data. The goal is to evaluate
how effectively a model’s output will transfer to a different collection of data. Preventing
overfitting, a situation in which a model learns the training data too well—including its
noise and outliers—and hence performs badly on unknown data, is one of the main goals
of cross-validation [71].
Water 2023, 15, 3970 25 of 37

4. Criterion for Evaluating Models

A measurement metric is a quantitative measure utilised to analyse the quality in
terms of efficiency or performance accuracy of a machine learning method. It gives a
standardised technique to evaluate how effectively the model functions concerning its
intended task or aim. In regression tasks, numerous evaluation metrics can be used to
analyse the performance of a regression model. Based on forecasting data and training
data, here are some indicators we utilised for our research result.

4.1. RMSE (Root Mean Squared Error)

RMSE is calculated by taking the square root of the average of the squared differences
between the predicted values (y_pred) and the actual values (y_true) (see Equations (17) and (18)).
√
RMSE = MSE (17)

1
q
N
RMSE =
N ∑ i
(y − ŷ)2
=1 i
(18)

where n is the number of data points and Σ represents the sum of squared differences
across all data points.

4.2. R-Squared (Coefficient of Determination)

R-squared (see Equation (19)) measures the proportion of variance in the dependent
variable (y_true) that is explained by the independent variables (y_pred). It ranges from 0
to 1, with 1 indicating a perfect fit.
2
∑(yi − ŷ)
R2 = 2
(19)
∑ ( yi − y )

where Σ represents the sum of squared differences between the predicted and actual values
and y_mean is the mean of the actual values.

5. Results
Figures 14–22 compare the actual rainfall with predicted rainfall based on temperature.
The graphs are plotted for both training data and testing data. The graph shows the data
distribution for both actual values during training and testing. The best results are obtained
using polynomial regression and random forest with an R2 value of 0.76. The RMSE values
are also shallow for these machine learning models. Also, the graphs for these models
(Figures 14 and 18) have similar distributions of actual and predicted values during testing,
whereas for other models, the distribution varies a lot between actual and predicted values.
The results describe the actual vs. predicted results of the models. The precipitation
recorded or witnessed at particular locations in Bangladesh over a specified period is
called actual rainfall. It is founded on information from satellite imaging, weather and
precipitation monitoring stations, and other reliable data sources.
The ground truth, or actual rainfall, is used to compare predictions and is essential for
evaluating the model’s accuracy. The estimations or forecasts of future rainfall produced
using a hydrological model, weather forecasting system, or predictive model are the
predicted rainfall. These forecasts are often based on historical weather data, atmospheric
conditions, and mathematical models that simulate precipitation patterns. In this work,
the variation in rainfall based on feature minimum or maximum temperature is measured
using multiple linear regression, polynomial linear regression, decision tree regression,
k-nearest neighbours, support vector machine, random forest, AdaBoostRegressor, Stacking
Regressor, and artificial neural network. Table 5 shows the results of model implementation.
 7. AdaBoostRegressor 0.7047 0.710915 110.9689 110.49437

8. Stacking Regressor 0.74631 0.738501 102.88535 106.1608

Artificial Neural Net-

Water 2023, 15, 3970 9. 0.763247 0.75847 100.911 26 of 100.77041
37
work

Water 2023, 15, x FOR PEER REVIEW 27

Water 2023, 15, x FOR PEER REVIEW 27 of

Figure 15. Illustration of actual and predicted rainfall using polynomial regression.
Figure 15. Illustration of actual and predicted rainfall using polynomial regression.

Figure 15. Illustration of actual and predicted rainfall using polynomial regression.

Figure 16. Cont.

 Water 2023, 15, x FOR PEER REVIEW 28 of 40
Water 2023, 15, x FOR PEER REVIEW 28 o
Water 2023, 15, 3970 27 of 37

Figure 16. Illustration of actual and predicted rainfall using decision tree.
Figure 16. Illustration of actual and predicted rainfall using decision tree.

Figure 16. Illustration of actual and predicted rainfall using decision tree.

Water 2023, 15, x FOR PEER REVIEW 29 of 40

Figure 17. Illustration of actual and predicted rainfall using k-nearest neighbour.
Figure 17. Illustration of actual and predicted rainfall using k-nearest neighbour.
Water 2023, 15, 3970 28 of 37
Figure 17. Illustration of actual and predicted rainfall using k-nearest neighbour.

Water 2023, 15, x FOR PEER REVIEW 30 of 40

Water 2023, 15, x FOR PEER REVIEW 30 of 40

Figure 18. Illustration of actual and predicted rainfall using support vector machine.
Figure 18. Illustration of actual and predicted rainfall using support vector machine.
Figure 18. Illustration of actual and predicted rainfall using support vector machine.

Figure 19. Cont.

 Water 2023, 15, x FOR PEER REVIEW 31

Water 2023, 15, x FOR PEER REVIEW 31 of 40

Water 2023, 15, 3970 29 of 37

Figure 19. Illustration of actual and predicted rainfall using random forest.
Figure 19. Illustration of actual and predicted rainfall using random forest.
Figure 19. Illustration of actual and predicted rainfall using random forest.

Water 2023, 15, x FOR PEER REVIEW 32 of

Figure 20. Illustration of actual and predicted rainfall using AdaBoost.

Figure 20. Illustration of actual and predicted rainfall using AdaBoost.
Water 2023, 15, 3970 30 of 37
Figure 20. Illustration of actual and predicted rainfall using AdaBoost.

Water 2023, 15, x FOR PEER REVIEW 33 of 40

Water 2023, 15, x FOR PEER REVIEW 33 of 40

Figure 21. Illustration of actual and predicted rainfall using the Stacking Regressor model.
Figure 21. Illustration of actual and predicted rainfall using the Stacking Regressor model.
Figure 21. Illustration of actual and predicted rainfall using the Stacking Regressor model.

Figure 22. Cont.

 Water 2023, 15, x FOR PEER REVIEW 34 of
Water 2023, 15, 3970 31 of 37

Figure 22. Illustration of actual and predicted rainfall using ANN.

Table 5. Results of implemented machine learning models.

Figure 22. Illustration of actual and predicted rainfall using ANN.

Machine and Deep Table 6 shows the results obtained via LSTM and RNN. Various statistics, such
Evaluation Metrics R2 and RMSE
Learning Model
loss, validation loss, RMSE and testing, are also shown. It can be seen that the loss valu
S.No. ML Model R2 Score Training
for LSTM areR2significantly
Score TestingbetterRMSE Score Training
than RNN. RMSE Testing
1. Multiple Linear regression 0.6643 0.6687 118.3231 118.279217
Table 6. Results obtained using various deep learning models (LSTM and RNN).
2. Polynomial regression 0.773177 0.7642164 99.12397 99.844
3. Decision Tree mode 0.75 0.72 101.195 123.27715
S. No Model Architecture Parameters Value
4. k-nearest neighbours 0.9992 0.74723 5.5840 103.31968
5. Support vector machine 0.654139 0.6583 120.108182
Loss 120.12110 0.0904
6. Random ForestLSTM 0.96417 0.768234 38.656 99.5790
In order to address and overcome the
7. AdaBoostRegressor 0.7047 0.710915 110.9689
RMSE 110.49437 0.3007
shortcomings of conventional RNNs,
8. Stacking1.Regressor 0.74631 0.738501
Refer to Figure 23 102.88535 106.1608
the LSTM approach was specifically
9. Artificial Neural Network 0.763247 0.75847 Val_loss
100.911 100.77041 0.0906
developed for learning long-term de-
pendencies.
Testing set loss 93,260.7188
Table 6 shows the results obtained via LSTM and RNN. Various statistics, such as loss,
Loss 126.5478
RNN
validation loss, RMSE and testing, are also shown. It can be seen that the loss values for
The main feature of an RNN is its abil-
LSTM are significantly better than RNN.
ity to maintain a hidden state or mean_absolute_error: 126.5478
memory, which is revised at each time
2 Refer to Figure 24
Table Results
6.passed
step and obtained
as input to the using
next, various deep learning models (LSTM and RNN).
Val_loss 124.1010
allowing the network to consider pre-
vious information while processing
S. No Model Architecture Parameters Value
the current input. Val_mean_absolute_error 124.1010
LSTM Loss 0.0904
In order to address and overcome the shortcomings of RMSE 0.3007
1. Refer to Figure 23
conventional RNNs, the LSTM approach was specifically Val_loss 0.0906
developed for learning long-term dependencies. Testing set loss 93,260.7188
RNN Loss 126.5478
The main feature of an RNN is its ability to maintain a
2. Refer to Figure 24 mean_absolute_error: 126.5478
hidden state or memory, which is revised at each time step
and passed as input to the next, allowing the network to Val_loss 124.1010
consider previous information while processing the
current input. Val_mean_absolute_error 124.1010
Water 2023, 15,
Water 2023, 15, 3970
x FOR PEER REVIEW 32of
35 of 37
40
Water 2023, 15, x FOR PEER REVIEW 35 of 40

Figure 23. Architecture of LSTM model used for training.

Figure 23.
Figure Architecture of
23. Architecture of LSTM
LSTM model
model used
used for
for training.
training.

Figure 24. Architecture of RNN model used for training.

Figure 24. Architecture of RNN model used for training.
Figure 24. Architecture of RNN model used for training.
6. Discussion
6. Discussion
This paper provided a comparison of various machine learning and deep learning
ThisItpaper
models. can beprovided
seen thatapolynomial
comparisonregression,
of variousrandommachineforestlearning
and and LSTM deep learning
provide the
models. It can be seen that polynomial regression, random forest
best results. We have primarily used sk-learn and TensorFlow along with k-fold cross- and LSTM provide the
best results.The
validation. We have
Weother primarily
havemachine
primarily used
used sk-learn
learning modelsand
sk-learn and TensorFlow
TensorFlow
employed along
along with
in previous with k-fold
studies gavecross-
k-fold cross-
poor
validation. The
The other machine learning models employed
employed in previous
results primarily because they failed to capture the complex trends available in the data. studies gave poor
results primarily
These models because
because
(linear they
they failed
regression, failed to
to capture
decision capture the complex
the support
trees and complexvectortrends available
available in
trendsmachines) in the
are the data.
data.
primar-
These models
models (linear
(linear regression,
regression, decision
decision trees
trees and
and support
support vector
vector
ily linear and cannot capture the non-linear trends in the data. Such models have shallow machines)
machines) are
areprimarily
primar-
linear
ily
capacity.andand
linear Incannot
cannot
contrast,capture
capture thethe
polynomial non-linear
non-linear
regression trends
trends ininthe
employs a data.
the data.Such
higher Such models
models
capacity modelhave
haveto shallow
capture
capacity.
capacity. In
In contrast,
contrast, polynomial
polynomial regression
regression employs
employs a higher
a higher
the variance in data. The hypothesis space of polynomial regression is much richer capacity
capacity modelmodelto capture
to capture
and the
it
variance
the in
variance data.
in The
data. hypothesis
The hypothesis space of
space polynomial
of polynomial regression
regression
can capture non-linear patterns in the data. The same goes for the random forest that usesis much
is richer
much and
richer it
and canit
capture
can in non-linear
treescapture to patterns
non-linear
parallel model in the
patterns
the indata.
complex The
datasame
the data. The goes goes
same
patterns. for the
forrandom
Finally, the random
LSTM isforest thatneural
forest
a deep uses
that trees
uses
net-
in parallel
trees to model the complex data patterns. Finally, LSTM is a deep neural network
work that has several layers along with the activation functions to model trends innet-
in parallel to model the complex data patterns. Finally, LSTM is a deep neural the
that has
work several
that has layers layers
several along withalong the activation
with the functions
activation to model
functions to trends
model intrends
the data.
in In
the
data. In addition, the LSTM model can capture the sequential patterns in the data. There-
addition,
data. the
In addition,LSTM model
the LSTM can capture
model can the
capturesequential
the patterns
sequential in the
patterns data.
in theTherefore,
data.studies
There-the
fore, the results obtained from these models are suitable. Finally, several similar
results
fore, theobtained
results from these
obtained models
from theseare suitable.
models areFinally,
suitable.several similar
Finally, several studies
similarhave been
studies
have been reported in the literature. Similar results have been reported in [71], 2 where R2
reported
have inreported
the literature. Similar results haveresults
been reported in [71], where inR[71],values were
valuesbeenwere obtained in to
thebeliterature. Similar
close to 0.87 and 0.92 [72]. have Thebeen reported
authors of [72] also where
reported R2
obtained
values to be
were close to to
obtained 0.87
be and
close 0.92
to [72].
0.87 The authors
and 0.92 [72].ofThe
[72]authors
also reported
of [72]similar
also results.
reported
similar results.
similar results.
Water 2023, 15, 3970 33 of 37

Before proceeding towards the end of the discussion, let us explore the results in depth.
Table 5 shows the results of various implemented machine learning models. Their R2 and
RMSE values are shown for both training and testing. The primary concern is the values
of these parameters during testing. A higher R2 value shows the supremacy of the model.
We can see that polynomial regression provides the highest R2 value of 0.76. Similarly, a
lower error values (RMSE) also shows the model is performing well. Again, the RMSE
value of 99.844 is the lowest for polynomial regression. There are other models as well,
such as multiple linear regression, decision tree, k-nearest neighbour and support vector
machines. However, these models have R2 values of less than 0.76. Figures 14–22 show the
actual and predicted rain fall with respect to temperature during both training and testing.
It can be seen that the trend/plot for polynomial regression is similar/closed for both
training and testing, whereas for other models, the plot differs significantly for training
and testing. Hence, we can say that polynomial regression provides a better modelling
of rainfall. Table 6 shows the results obtained via deep learning models. The loss and
RMSE values for LSTM are significantly better than those obtained for RNN because LSTM
captures long-term dependencies and RNN suffers from vanishing gradient and exploding
gradient problems. Therefore, LSTM has performed with better results.
There are certain limitations in this work. The dataset was limited, and we have only
tested a few deep learning models. More work can be carried out to employ pre-trained
models and transfer learning to improve performance. We have only evaluated this model
for specific parameters. Other parameters can also be considered for extensive evaluations.

7. Conclusions and Future Work

The development of a machine-learning-based rain prediction model is the aim of
this study. The model used is based on a dataset of 2391 observations that the Bangladesh
Meteorological Department (BMD) in Dhaka has gathered. We combined nine machine
learning models and two deep learning algorithms for this research. Before the rainfall
forecasts were verified, every model was trained using 16 input characteristics. The
simulation’s performance outcomes were all satisfactory. Future versions of this model
will have a system for early warning and include additional parameters like humidity,
wind gusts, pressure in the air, atmospheric pressure, sun exposure to radiation, etc. This
will enable the early detection of natural disasters and enhance real-time global warming
predictions. The difficulty in guessing which models are most appealing or how many days
in advance are ideal for making a prediction would be another highly intriguing future
study related to the one just mentioned. It is also planned to test the approach for other
countries along with state-of-the-art machine learning models. One can also develop a
disaster response system once a disaster has happened.
Finally, data from several meteorological departments in various nations or areas
may be included in future research. These would offer a broader range of information,
capturing various climatic trends and improving the model’s suitability. Hydrological
model integration, in addition to mathematical flood channel modelling, may provide a
better understanding of land–water interactions and increase the accuracy of flood forecasts.
In the near future, researchers can look at the prospect of adding additional characteristics
that can have a significant impact on flooding, such as soil moisture, changes in land use,
or urbanisation measures. Furthermore, temporal patterns in greater detail while taking
seasonality and long-term climate change into account could be examined. This could lead
to better performance from the LSTM model or indicate the need for further temporal-
based neural network topologies. Future researchers should examine how man-made
constructions like levees, reservoirs, and dams affect flood forecasting. Predictions in areas
with substantial human involvement may become more accurate due to understanding
these interventions.
Moreover, historical accounts and first-hand experience from flood-prone communities
might offer insightful information. Subsequent studies can use input from the community
or personal experiences to improve further prediction algorithms. Further research could
Water 2023, 15, 3970 34 of 37

concentrate on creating a complete end-to-end early warning system that combines predic-
tion with communication channels to notify locals in flood-prone areas, building on the
predictive skills.

Author Contributions: Conceptualisation, A.R., H.F., N.I. and D.S.; methodology, M.A.E., A.S., M.A.
(Muhammad Akram) and M.A. (Mesfer Alrizq); software, A.R., H.F., N.I. and D.S.; validation, M.A.E.,
A.S., M.A. (Muhammad Akram). and M.A. (Mesfer Alrizq); formal analysis, A.R., H.F., N.I. and D.S.;
investigation, M.A.E., A.S., M.A. (Muhammad Akram) and M.A. (Mesfer Alrizq); resources, H.F.
and N.I.; data curation, M.A.E. and M.A. (Mesfer Alrizq); writing—original draft preparation, A.R.,
H.F., N.I. and D.S.; writing—review and editing, M.A.E., A.S., M.A. (Muhammad Akram). and M.A.
(Mesfer Alrizq); visualisation, M.A. (Muhammad Akram).; supervision, A.S.; project administration,
A.R. and N.I.; funding acquisition, M.A.E. All authors have read and agreed to the published version
of the manuscript.
Funding: The authors would like to acknowledge the support of the Deputy for Research and
Innovation, Ministry of Education, Kingdom of Saudi Arabia, for funding this research through a
grant (NU/IFC/2/SERC/-/48) under the Institutional Funding Committee at Najran University,
Kingdom of Saudi Arabia.
Data Availability Statement: Data available in a publicly accessible repository on Kaggle and can be
found at the following link: https://www.kaggle.com/datasets/emonreza/65-years-of-weather-
data-bangladesh-preprocessed, accessed on 20 October 2023.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Syeed, M.M.A.; Farzana, M.; Namir, I.; Ishrar, I.; Nushra, M.H.; Rahman, T. Flood prediction using machine learning models.
In Proceedings of the 2022 International Congress on Human-Computer Interaction, Optimization and Robotic Applications
(HORA), Ankara, Turkey, 9–11 June 2022; IEEE: New York, NY, USA, 2022.
2. Kumar, V.; Azamathulla, H.M.; Sharma, K.V.; Mehta, D.J.; Maharaj, K.T. The state of the art in deep learning applications,
challenges, and future prospects: A comprehensive review of flood forecasting and management. Sustainability 2023, 15, 10543.
[CrossRef]
3. Gude, V.; Corns, S.; Long, S. Flood prediction and uncertainty estimation using deep learning. Water 2020, 12, 884. [CrossRef]
4. Nguyen, D.T.; Chen, S.-T. Real-time probabilistic flood forecasting using multiple machine learning methods. Water 2020, 12, 787.
[CrossRef]
5. Furquim, G.; Pessin, G.; Faiçal, B.S.; Mendiondo, E.M.; Ueyama, J. Improving the accuracy of a flood forecasting model by means
of machine learning and chaos theory: A case study involving a real wireless sensor network deployment in brazil. Neural
Comput. Appl. 2016, 27, 1129–1141. [CrossRef]
6. Talukdar, S.; Ghose, B.; Shahfahad; Salam, R.; Mahato, S.; Pham, Q.B.; Linh, N.T.T.; Costache, R.; Avand, M. Flood susceptibility
modeling in Teesta River basin, Bangladesh using novel ensembles of bagging algorithms. Stoch. Environ. Res. Risk Assess. 2020,
34, 2277–2300. [CrossRef]
7. Maspo, N.-A.; Bin Harun, A.N.; Goto, M.; Cheros, F.; Haron, N.A.; Nawi, M.N.M. Evaluation of Machine Learning approach in
flood prediction scenarios and its input parameters: A systematic review. In IOP Conference Series: Earth and Environmental Science;
IOP Publishing: Bristol, UK, 2020.
8. Mitra, P.; Ray, R.; Chatterjee, R.; Basu, R.; Saha, P.; Raha, S.; Barman, R.; Patra, S.; Biswas, S.S.; Saha, S. Flood forecasting
using Internet of things and artificial neural networks. In Proceedings of the 2016 IEEE 7th Annual Information Technology,
Electronics and Mobile Communication Conference (IEMCON), Vancouver, BC, Canada, 13–15 October 2016; IEEE: New York,
NY, USA, 2016.
9. Noymanee, J.; Nikitin, N.O.; Kalyuzhnaya, A.V. Urban pluvial flood forecasting using open data with machine learning techniques
in pattani basin. Procedia Comput. Sci. 2017, 119, 288–297. [CrossRef]
10. Aswad, F.M.; Kareem, A.N.; Khudhur, A.M.; Khalaf, B.A.; Mostafa, S.A. Tree-based machine learning algorithms in the Internet of
Things environment for multivariate flood status prediction. J. Intell. Syst. 2021, 31, 1–14. [CrossRef]
11. Sankaranarayanan, S.; Prabhakar, M.; Satish, S.; Jain, P.; Ramprasad, A.; Krishnan, A. Flood prediction based on weather
parameters using deep learning. J. Water Clim. Change 2020, 11, 1766–1783. [CrossRef]
12. Wang, G.; Yang, J.; Hu, Y.; Li, J.; Yin, Z. Application of a novel artificial neural network model in flood forecasting. Environ. Monit.
Assess. 2022, 194, 125. [CrossRef]
13. Puttinaovarat, S.; Horkaew, P. Flood forecasting system based on integrated big and crowdsource data by using machine learning
techniques. IEEE Access 2020, 8, 5885–5905. [CrossRef]
Water 2023, 15, 3970 35 of 37

14. Ria, N.J.; Ani, J.F.; Islam, M.; Masum, A.K.M. Standardization Of Rainfall Prediction In Bangladesh Using Machine Learning
Approach. In Proceedings of the 2021 12th International Conference on Computing Communication and Networking Technologies
(ICCCNT), Kharagpur, India, 6–8 July 2021; IEEE: New York, NY, USA, 2021.
15. Osmani, S.A.; Kim, J.-S.; Jun, C.; Sumon, W.; Baik, J.; Lee, J. Prediction of monthly dry days with machine learning algorithms: A
case study in Northern Bangladesh. Sci. Rep. 2022, 12, 19717. [CrossRef] [PubMed]
16. Manandhar, A.; Fischer, A.; Bradley, D.J.; Salehin, M.; Islam, M.S.; Hope, R.; Clifton, D.A. Machine learning to evaluate impacts of
flood protection in Bangladesh, 1983–2014. Water 2020, 12, 483. [CrossRef]
17. Aydin, M.C.; Sevgi Birincioğlu, E. Flood risk analysis using gis-based analytical hierarchy process: A case study of Bitlis Province.
Appl. Water Sci. 2022, 12, 122. [CrossRef]
18. Msabi, M.M.; Makonyo, M. Flood susceptibility mapping using GIS and multi-criteria decision analysis: A case of Dodoma
region, central Tanzania. Remote Sens. Appl. Soc. Environ. 2021, 21, 100445. [CrossRef]
19. Shafizadeh-Moghadam, H.; Valavi, R.; Shahabi, H.; Chapi, K.; Shirzadi, A. Novel forecasting approaches using combination of
machine learning and statistical models for flood susceptibility mapping. J. Environ. Manag. 2018, 217, 1–11. [CrossRef]
20. Elmagzoub, M.; Syed, D.; Shaikh, A.; Islam, N.; Alghamdi, A.; Rizwan, S. A survey of swarm intelligence based load balancing
techniques in cloud computing environment. Electronics 2021, 10, 2718. [CrossRef]
21. Al Reshan, M.S.; Syed, D.; Islam, N.; Shaikh, A.; Hamdi, M.; Elmagzoub, M.A.; Muhammad, G.; Talpur, K.H. A Fast Converging
and Globally Optimized Approach for Load Balancing in Cloud Computing. IEEE Access 2023, 11, 11390–11404. [CrossRef]
22. Islam, N.; Raza, E.; Mohsin, S.; Ansari, A.; Shuja, R.; Syed, D. Forecasting on COVID-19 Data Using ARIMAX Model. In Data
Science with Semantic Technologies; CRC Press: Boca Raton, FL, USA, 2023; pp. 95–113.
23. Islam, N.; Khan, S.K.; Rehman, A.; Aftab, U.; Syed, D. Stock Prediction for ARGAAM Companies Dataset. KIET J. Comput. Inf. Sci.
2023, 6, 1–13. [CrossRef]
24. Bui, D.T.; Pradhan, B.; Nampak, H.; Bui, Q.-T.; Tran, Q.-A.; Nguyen, Q.-P. Hybrid artificial intelligence approach based on neural
fuzzy inference model and metaheuristic optimization for flood susceptibilitgy modeling in a high-frequency tropical cyclone
area using GIS. J. Hydrol. 2016, 540, 317–330.
25. Chatterjee, S.; Datta, B.; Sen, S.; Dey, N.; Debnath, N.C. Rainfall prediction using hybrid neural network approach. In Proceed-
ings of the 2018 2nd International Conference on Recent Advances in Signal Processing, Telecommunications & Computing
(SigTelCom), Ho Chi Minh, Vietnam, 29–31 January 2018; IEEE: New York, NY, USA, 2018.
26. Islam, M.N.; van Amstel, A.; Ghosh, B.K.; Sarker, K.R. Climate Change and Living with Floods: An Empirical Case from the
Saghata Union of Gaibandha District, Bangladesh. In Bangladesh II: Climate Change Impacts, Mitigation and Adaptation in Developing
Countries; Springer: Cham, Switzerland, 2021; pp. 459–478.
27. Luo, T.; Maddocks, A.; Iceland, C.; Ward, P.; Winsemius, H. World’s 15 Countries with the Most People Exposed to River Floods; World
Resources Institute: Washington, DC, USA, 2015.
28. Kumari, S.; Tripathy, K.K.; Kumbhar, V. Data Science and Analytics; Emerald Publishing Limited: Bingley, UK, 2020.
29. Thirumalai, C.; Harsha, K.S.; Deepak, M.L.; Krishna, K.C. Heuristic prediction of rainfall using machine learning techniques. In
Proceedings of the 2017 International Conference on Trends in Electronics and Informatics (ICEI), Tirunelveli, India, 11–12 May
2017; IEEE: New York, NY, USA, 2017.
30. Adnan, R.; Zain, Z.M.; Ruslan, F.A. 5 hours flood prediction modeling using improved NNARX structure: Case study Kuala
Lumpur. In Proceedings of the 2014 IEEE 4th International Conference on System Engineering and Technology (ICSET), Bandung,
Indonesia, 24–25 November 2014; IEEE: New York, NY, USA, 2014.
31. Mosavi, A.; Ozturk, P.; Chau, K.-W. Flood prediction using machine learning models: Literature review. Water 2018, 10, 1536.
[CrossRef]
32. Chen, C.; Jiang, J.; Liao, Z.; Zhou, Y.; Wang, H.; Pei, Q. A short-term flood prediction based on spatial deep learning network: A
case study for Xi County, China. J. Hydrol. 2022, 607, 127535. [CrossRef]
33. Motta, M.; de Castro Neto, M.; Sarmento, P. A mixed approach for urban flood prediction using Machine Learning and GIS. Int. J.
Disaster Risk Reduct. 2021, 56, 102154. [CrossRef]
34. Ghorpade, P.; Gadge, A.; Lende, A.; Chordiya, H.; Gosavi, G.; Mishra, A.; Hooli, B.; Ingle, Y.S.; Shaikh, N. Flood forecasting using
machine learning: A review. In Proceedings of the 2021 8th International Conference on Smart Computing and Communications
(ICSCC), Kerala, India, 1–3 July 2021; IEEE: New York, NY, USA, 2021.
35. Adnan, M.S.G.; Siam, Z.S.; Kabir, I.; Kabir, Z.; Ahmed, M.R.; Hassan, Q.K.; Rahman, R.M.; Dewan, A. A novel framework
for addressing uncertainties in machine learning-based geospatial approaches for flood prediction. J. Environ. Manag. 2023,
326, 116813. [CrossRef] [PubMed]
36. Gauhar, N.; Das, S.; Moury, K.S. Prediction of flood in Bangladesh using K-nearest neighbors algorithm. In Proceedings of the
2021 2nd International Conference on Robotics, Electrical and Signal Processing Techniques (ICREST), Dhaka, Bangladesh, 5–7
January 2021; IEEE: New York, NY, USA, 2021.
37. Han, S.; Coulibaly, P. Bayesian flood forecasting methods: A review. J. Hydrol. 2017, 551, 340–351. [CrossRef]
38. Hamidul Haque, M.; Sadia, M.; Mustaq, M. Development of Flood Forecasting System for Someshwari-Kangsa Sub-watershed of
Bangladesh-India Using Different Machine Learning Techniques. EGU General Assembly Conference Abstracts; EGU: Virtual, 2021.
Available online: https://ui.adsabs.harvard.edu/abs/2021EGUGA..2315294H/abstract (accessed on 20 October 2023).
Water 2023, 15, 3970 36 of 37

39. Billah, M.; Adnan, N.; Akhond, M.R.; Ema, R.R.; Hossain, A.; Galib, S.M. Rainfall prediction system for Bangladesh using long
short-term memory. Open Comput. Sci. 2022, 12, 323–331. [CrossRef]
40. Yaseen, M.W.; Awais, M.; Riaz, K.; Rasheed, M.B.; Waqar, M.; Rasheed, S. Artificial Intelligence Based Flood Forecasting for River
Hunza at Danyor Station in Pakistan. Arch. Hydro-Eng. Environ. Mech. 2022, 69, 59–77. [CrossRef]
41. Parmar, A.; Mistree, K.; Sompura, M. Machine learning techniques for rainfall prediction: A review. In Proceedings of the
International Conference on Innovations in Information Embedded and Communication Systems, Coimbatore, India, 17–18
March 2017.
42. Khosravi, K.; Panahi, M.; Golkarian, A.; Keesstra, S.D.; Saco, P.M.; Bui, D.T.; Lee, S. Convolutional neural network approach for
spatial prediction of flood hazard at national scale of Iran. J. Hydrol. 2020, 591, 125552. [CrossRef]
43. Kovalchuk, S.V.; Krikunov, A.V.; Knyazkov, K.V.; Boukhanovsky, A.V. Classification issues within ensemble-based simulation:
Application to surge floods forecasting. Stoch. Environ. Res. Risk Assess. 2017, 31, 1183–1197. [CrossRef]
44. Nevo, S.; Morin, E.; Rosenthal, A.G.; Metzger, A.; Barshai, C.; Weitzner, D.; Voloshin, D.; Kratzert, F.; Elidan, G.; Dror, G.; et al.
Flood forecasting with machine learning models in an operational framework. arXiv 2021, arXiv:2111.02780. [CrossRef]
45. Khosravi, K.; Shahabi, H.; Pham, B.T.; Adamowski, J.; Shirzadi, A.; Pradhan, B.; Dou, J.; Ly, H.-B.; Gróf, G.; Ho, H.L.; et al. A
comparative assessment of flood susceptibility modeling using multi-criteria decision-making analysis and machine learning
methods. J. Hydrol. 2019, 573, 311–323. [CrossRef]
46. El-Magd, S.A.A.; Pradhan, B.; Alamri, A. Machine learning algorithm for flash flood prediction mapping in Wadi El-Laqeita and
surroundings, Central Eastern Desert, Egypt. Arab. J. Geosci. 2021, 14, 323. [CrossRef]
47. Nayak, M.; Das, S.; Senapati, M.R. Improving Flood Prediction with Deep Learning Methods. J. Inst. Eng. Ser. B 2022, 103,
1189–1205. [CrossRef]
48. Tayfur, G.; Singh, V.P.; Moramarco, T.; Barbetta, S. Flood hydrograph prediction using machine learning methods. Water 2018,
10, 968. [CrossRef]
49. Sahoo, A.; Samantaray, S.; Ghose, D.K. Prediction of flood in Barak River using hybrid machine learning approaches: A case
study. J. Geol. Soc. India 2021, 97, 186–198. [CrossRef]
50. Qian, K.; Mohamed, A.; Claudel, C. Physics informed data driven model for flood prediction: Application of deep learning in
prediction of urban flood development. arXiv 2019, arXiv:1908.10312.
51. Miau, S.; Hung, W.-H. River flooding forecasting and anomaly detection based on deep learning. IEEE Access 2020, 8,
198384–198402. [CrossRef]
52. Hossain, I.; Rasel, H.M.; Alam Imteaz, M.; Mekanik, F. Long-term seasonal rainfall forecasting using linear and non-linear
modelling approaches: A case study for Western Australia. Meteorol. Atmos. Phys. 2020, 132, 131–141. [CrossRef]
53. Ighile, E.H.; Shirakawa, H.; Tanikawa, H. Application of GIS and machine learning to predict flood areas in Nigeria. Sustainability
2022, 14, 5039. [CrossRef]
54. Kunverji, K.; Shah, K.; Shah, N. A flood prediction system developed using various machine learning algorithms. In Proceedings
of the 4th International Conference on Advances in Science & Technology (ICAST2021), Mumbai, India, 7 May 2021.
55. Dodangeh, E.; Choubin, B.; Eigdir, A.N.; Nabipour, N.; Panahi, M.; Shamshirband, S.; Mosavi, A. Integrated machine learning
methods with resampling algorithms for flood susceptibility prediction. Sci. Total Environ. 2020, 705, 135983. [CrossRef]
56. Khairudin, N.M.; Mustapha, N.O.; Aris, T.N.; Zolkepli, M.A. A study to investigate the effect of different time-series scales
towards flood forecasting using machine learning. J. Theor. Appl. Inform. Technol. 2021, 99, 5687–5699.
57. Dtissibe, F.Y.; Ari, A.A.A.; Titouna, C.; Thiare, O.; Gueroui, A.M. Flood forecasting based on an artificial neural network scheme.
Nat. Hazards 2020, 104, 1211–1237. [CrossRef]
58. Sarasa-Cabezuelo, A. Prediction of rainfall in Australia using machine learning. Information 2022, 13, 163. [CrossRef]
59. Liyew, C.M.; Melese, H.A. Machine learning techniques to predict daily rainfall amount. J. Big Data 2021, 8, 153. [CrossRef]
60. Singh, P. Indian summer monsoon rainfall (ISMR) forecasting using time series data: A fuzzy-entropy-neuro based expert system.
Geosci. Front. 2018, 9, 1243–1257. [CrossRef]
61. Mishra, N.; Soni, H.K.; Sharma, S.; Upadhyay, A.K. Development and analysis of artificial neural network models for rainfall
prediction by using time-series data. Int. J. Intell. Syst. Appl. 2018, 12, 16. [CrossRef]
62. Chitwatkulsiri, D.; Miyamoto, H. Real-Time Urban Flood Forecasting Systems for Southeast Asia—A Review of Present Modelling
and Its Future Prospects. Water 2023, 15, 178. [CrossRef]
63. Kumar, V.; Sharma, K.V.; Caloiero, T.; Mehta, D.J.; Singh, K. Comprehensive overview of flood modeling approaches: A review of
recent advances. Hydrology 2023, 10, 141. [CrossRef]
64. Mosaffa, H.; Sadeghi, M.; Mallakpour, I.; Jahromi, M.N.; Pourghasemi, H.R. Application of Machine Learning Algorithms in
Hydrology. In Computers in Earth and Environmental Sciences; Elsevier: Amsterdam, The Netherlands, 2022; pp. 585–591.
65. Jehanzaib, M.; Ajmal, M.; Achite, M.; Kim, T.-W. Comprehensive review: Advancements in rainfall-runoff modelling for flood
mitigation. Climate 2022, 10, 147. [CrossRef]
66. Mistry, S.; Parekh, F. Flood Forecasting Using Artificial Neural Network. In IOP Conference Series: Earth and Environmental Science;
IOP Publishing: Bristol, UK, 2022.
67. Xu, Y.; Hu, C.; Wu, Q.; Jian, S.; Li, Z.; Chen, Y.; Zhang, G.; Zhang, Z.; Wang, S. Research on particle swarm optimization in LSTM
neural networks for rainfall-runoff simulation. J. Hydrol. 2022, 608, 127553. [CrossRef]
Water 2023, 15, 3970 37 of 37

68. Cho, M.; Kim, C.; Jung, K.; Jung, H. Water level prediction model applying a long short-term memory (lstm)–gated recurrent unit
(gru) method for flood prediction. Water 2022, 14, 2221. [CrossRef]
69. Qadeer, K.; Rehman, W.U.; Sheri, A.M.; Park, I.; Kim, H.K.; Jeon, M. A long short-term memory (LSTM) network for hourly
estimation of PM2.5 concentration in two cities of South Korea. Appl. Sci. 2020, 10, 3984. [CrossRef]
70. Available online: https://www.kaggle.com/datasets/emonreza/65-years-of-weather-data-bangladesh-preprocessed (accessed
on 20 October 2023).
71. Wong, T.-T.; Yeh, P.-Y. Reliable accuracy estimates from k-fold cross validation. IEEE Trans. Knowl. Data Eng. 2019, 32, 1586–1594.
[CrossRef]
72. Rahman, M.; Chen, N.; Elbeltagi, A.; Islam, M.M.; Alam, M.; Pourghasemi, H.R.; Tao, W.; Zhang, J.; Shufeng, T.; Faiz, H.; et al.
Application of stacking hybrid machine learning algorithms in delineating multi-type flooding in Bangladesh. J. Environ. Manag.
2021, 295, 113086. [CrossRef] [PubMed]

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.