See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/382339588

AI- and IoT-Based Applications for Rainfall Prediction

Chapter · July 2024

DOI: 10.1201/9781003452393-3

CITATIONS READS
0 108

6 authors, including:

Manoj Kumar Pandey Jyoti Upadhyay

Chandigarh University Rungta Group of Colleges
25 PUBLICATIONS 51 CITATIONS 11 PUBLICATIONS 5 CITATIONS

SEE PROFILE SEE PROFILE

Preeti Tiwari Naresh Kumar Kar

IIT Bhilai innovation and Technology Foundation GITAM University
3 PUBLICATIONS 1 CITATION 7 PUBLICATIONS 4 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Manoj Kumar Pandey on 10 August 2024.

The user has requested enhancement of the downloaded file.

 AI and IoT based application for rainfall prediction: A Study

Manoj Kumar Pandey 1*

1*
Assistant professor, Pranveer Singh Institute of Technology, Kanpur
1*
[email protected]
Sunil Kumar Singh 2
2
Assistant Professor, CMREC, Hyderabad
2
[email protected]
Jyoti Upadhyay3
3
Assistant Professor, GD Rungta College of Science and Technology, Bhilai
3
[email protected]
Preeti Tiwari 4
4
Project Associate, IIT, Bhilai
4
[email protected]
Naresh Kumar Kar 5
5
Assistant Professor, GITAM Deemed to be University, Hyderabad
5
[email protected]
Jai Prakash Kushwaha 6
6
Research Scholar, ABV-IIITM, Gwalior
6
[email protected]

Introduction: Rainfall prediction is one of the important aspects for everyone and usages of
artificial intelligence (AI) and Internet of Things (IoT) may impact the overall process of prediction.
Rainfall prediction is very important for everyone for several of reasons and making good prediction
impacts the agriculture in a greater extent. The scarcity of rainfall has a negative impact on the
aquatic ecosystem, water supply and productivity. The rainfall prediction is based on various factors
such as wind speed, moisture, humidity, temperature; pressure, min-temp etc., and theses parameters
can be very useful to predict the rainfall by applying various machine learning models. These
predictions are based on the datasets which belongs to regions and bounds to time duration and by
applying this dataset rainfall is predicted by applying various machine learning models. The time
duration of data collection also affects the overall performance of rainfall prediction, assume what if
one can get the real time data for training purpose, it will improve the model performance and
prediction can be effective and accurate. IoT is a field which enables all of us to collect the real time
data using various sensors used for specialized task. These sensors are installed and are well
connected with the cloud servers so that real time data can be collected easily for good predictions.
This book chapter contains various aspect of rainfall prediction using AI and IoT.

A transformative era has recently begun as a result of the convergence of Artificial Intelligence (AI)
and the Internet of Things (IoT), which has completely changed how data is gathered, examined, and
applied to solve complicated problems. Predicting rainfall is a crucial field with broad consequences
for disaster planning, agricultural production, water resource management, and other issues. This
synergy has several intriguing applications in this field. This dynamic combination of AI and IoT
technologies has made it possible to create complex systems that can provide accurate forecasts in a
timely manner, improving our capacity to foresee and react to changes in precipitation patterns
[1].The development of artificial intelligence has significantly altered how we examine enormous,
complex datasets, revealing patterns and links that were previously hidden. AI approaches enable
the extraction of useful insights from a wide range of meteorological variables, historical data, and
real-time observations collected by IoT devices in the context of rainfall prediction. Few machine
learning algorithms such as decision trees [2], K- nearest neighbour [3], support vector machines,
and neural networks [4] are examples of machine learning algorithms that independently learn from
previous data and continuously improve their predictions through iterative learning processes. As a
result, a prediction framework that can recognise the complex, nonlinear interactions that frequently
control rainfall patterns is created. Arabelli et al., 2023, [5] has employed decision trees (DT),
logistic regression (LR), support vector machine (SVM) and random forests (RF) to perform the
weather prediction and for collecting real time data various sensors like DHT11, and BMP180 is
used and one module ESP8266 is used for the add the sensor generated data to the database. Here
the random forest models show the highest accuracy of 84% and takes time of 2.4 second.

It is obvious that recent advancements in the IoT space have given us all the power to gather real-
time information on the forecast for rainfall, and using that information to make a prediction using a
machine learning approach can provide some truly amazing outcomes. The datasets, pre-processing
method, and models chosen for rainfall prediction all play a major role in rainfall forecast. To
effectively plan for the country and to ensure that standards of life are maintained, it is crucial to
anticipate rainfall with great accuracy. Predicting rainfall has many benefits for nations, benefiting
different societal sectors and facets. Its contribution to agricultural planning and food security is a
notable advantage.. Accurate rainfall forecasts enable farmers to make informed decisions about
planting, irrigation, and harvesting, optimizing crop yields, and ensuring a stable food supply for the
population. With the aid of precise rainfall projections, farmers may make knowledgeable choices
about planting, irrigation, and harvesting to maximise crop yields and guarantee a consistent supply
of food for the populace. [6].

Additionally, rainfall prediction plays a pivotal role in water resource management. By anticipating
precipitation patterns, authorities can effectively manage reservoir levels, groundwater recharge, and
water distribution for various uses such as domestic consumption, agriculture, and industrial
processes. This contributes to sustainable water management and helps mitigate the impact of water
scarcity [7]. By offering early warning systems for catastrophic weather occurrences like floods and
landslides, the projections also improve disaster preparedness. Initiating evacuation preparations,
allocating resources, and coordinating emergency response activities are all made possible by
prompt alerts based on rainfall projections, which lowers the loss of life and damage.

Accurate rainfall forecasting is important for infrastructure planning. With the help of projected
rainfall patterns, urban planners and engineers may create robust drainage systems and flood
management plans. By ensuring infrastructure longevity and lowering the risk of urban floods, these
measures eventually lead to lower maintenance costs [8]. The ability to predict rainfall is crucial for
energy generation, particularly in hydroelectric power plants. Power generation is directly impacted
by the availability of water in reservoirs. Energy companies can optimise their operations, control
reservoir levels, and guarantee a reliable and efficient energy supply with the help of precise
forecasts [9].

Forecasting rainfall benefits economic stability across a range of industries. Businesses in industry,
tourism, and agriculture can modify their plans based on the forecasted weather, reducing losses and
maximising income [10]. Prediction of rainfall has health effects as well. Authorities can take action
to prevent water stagnation and the creation of breeding grounds for disease-carrying insects like
mosquitoes by planning beforehand for heavy rain. By doing so, public health is protected and
vector-borne diseases are controlled. Insights gained from precise rainfall forecasting also aid in the
study and modelling of the climate [11].

2. Related Work

This section of chapter deals with the recent studies done for predicting the rain fall using IoT
devices and machine learning methods. The purpose of this section is to make acquaint reader about
the current trends in this direction of the research. After looking at various review of literature it is
observed that there are lots of scholarly articles related to IoT and rainfall, but here articles of last 6-
7 years have been considered so that this study can deal with some of the recent significant work.

Shah et al. 2023 [12] have done a study to predict the rain fall of three regions i.e., Chennai, Calicut,
Vadodara. In order to collect the latest rain-fall data, IoT devices have been employed and kept in
the three mentioned city. IoT devices DHT 11 and BMP 180 have been controlled using
microcontroller ESP8266 and interfaced with the PowerBi so that collected data can be easily
managing and shared. Regression and classification algorithms are used to predict the rain-fall in the
city. Decision tree regression and random forest regression algorithm have been used for the
regression analysis and decision tree classifier and naïve bayes classifiers have been used for the
classification. Along with the dataset collected using the IoT devices another standard database has
been used and performance of various machine learning models using these two databases have been
compared and it is found that decision tree regression model gives an accuracy of 84.73% for raw
data and accuracy of 88.23% for IoT based dataset. Likewise for random forest regression it is
showing accuracy of 85.20% for Raw data and accuracy of 85.68% for IoT based dataset. For the
classification task decision tree classifier achieves an accuracy of 92% for raw data and accuracy of
95% for IoT based dataset. In case of naïve bayes classifier it achieves an accuracy of 81% for raw
data and accuracy of 83% for the IoT based dataset. The above result and the conclusion of the study
shows that there is a difference of 2-3% accuracy, i.e., IoT based dataset gives an improved accuracy
of 2-3% as compared to raw dataset-based accuracy. It also shows that using IoT for recording the
data related to rain-fall may lead to the good result.

Liyew et al. 2021 [13] conducted yet another investigation to pinpoint the pertinent atmospheric
elements that cause rain to fall and to forecast the daily severity of rail-fall using machine learning
techniques. Three different machine learning models, including multivariate linear regression
(MLR), random forest (RF), and extreme gradient boost (XGBoost), have been used to collect data
from Ethiopia's local meteorological agency. For the purposes of the study, a 20-year-old raw
dataset containing ten data features—including year, month, date, evaporation, daylight, maximum
temperature, minimum temperature, humidity, wind speed, and rain fall—was taken into
consideration. Mean value replacement has been used to deal with the missing data. The entire
datasets have been partitioned into the 80% training and 20% testing. Pearson correlation is also
applied to find the significant features for rain prediction and it is found that features humidity and
wind speed are the significant feature for rain fall prediction having Pearson coefficient value of
0.402 and 0.351 (greater than 0.2). The rain-fall prediction is performed using the all three
mentioned machine learning algorithm and it is found that XGBoost predict the rain-fall better than
the RF and MLR model. The author concluded that this result may me further improved if sensor
data is incorporated for the study.

In order to make accurate predictions, Sathya et al. 2023 [14] suggested an Enhanced Learning
Scheme for Weather Prediction (ELSWP), which is based on the traditional machine learning
Logistic Regression (LR) model and used IoT devices DHT-11 and rain fall sensor. However, it has
a drawback in that it can occasionally function poorly because the data collected by IoT devices only
covers a small area.

Sharma, et al., 2022[15] have presented a study for the forecast of rainfall for farmers, and for that
purpose a 'SMART CAP' is built, which is wearable and comprise of microprocessor that can
capture atmosphere parameter with the assistance of several sensors. It can be worn by farmers as
they labour so they can record the temperature and humidity. With the aid of a WIFI module, the
recorded value is then entered into "THINGSPEAK" online. The prediction accuracy using the
linear and logistic regression models is 81% and 80.23%, respectively.

Maliyeckel et al. 2021 [16] have done a study to measure the performance of machine learning
algorithms like light gradient-boosting machine (LightGBM) and support vector regression (SVR)
and they have also developed an ensemble model based on LightGBM and SVR to predict the
rainfall. In order to accomplish this study they have utilized the database having three features i.e.,
humidity, wind-speed, and temperature. Rain-fall database from the period of 1st January 2019 to
31st August 2020 is considered for the training and database from the period of 1st September 2020
to 30 September 2020 is considered as the testing dataset. Missing data is managed using the mean
value. Pearson correlation coefficient is used to find the linear relation between the two variables
and after evaluation it is found that all the parameters are significantly important to rainfall. The
mean absolute error (MAE) evaluated for the LightGBM is 5.73, and for SVR MAE are 8.78.
However, the hybrid model shows an MAE of 4.90, which is more efficient than the other classical
model.

Another study, by Dash et al. 2018 [17], used three machine learning techniques— Artificial Neural
Network (ANN), K-nearest neighbour (KNN) and Extreme Learning Machine (ELM)—to estimate
rainfall in the Kerala state from 2011 to 2016. The last 47 years of data, separated into the months of
January–February, March–May, June-September, and October–December, are used to train the
machine learning models. Every piece of training data contains several statistical metrics, such as
mean, median, mode, lowest, and maximum. System performance is gauged using Root Mean
Square Error (RMSE), Mean Absolute Error (MAE), Standard Deviation (SD), and Mean. The
addition of the hidden layer has a greater impact on the outcome for ANN and ELM, and the 8-15-1
ELM architecture exhibits positive outcomes. The experimental finding indicates that, with MAE of
3.075% for summer monsoon datasets and MAE of 3.149% for post monsoon datasets, the ELM
technique has outperformed the other two strategies.

Another study done for rain-fall prediction is done by Huang et al. 2017 [18], which are totally
based on the K-Nearest neighbour algorithm. In this work improved KNN is utilized for rain
prediction over three other methods namely distance weighted K nearest neighbour (WKNN) and
dual weighted K nearest neighbour (DWKNN), KNN. Here the training sample used is 1748 record
and testing samples contains the 368 records. The proposed improved KNN shows the better
performance score of 50.46% for K=6 against the three other algorithm namely DWNN, WKNN
and KNN with scores of 48.32%, 48.62% and 46.88%.

In a study conducted by Chao et al. in 2018 [19], rainfall was predicted utilizing a micro-
electromechanical systems (MEMS) sensor to store relevant data for waterfall prediction and a long
short-term memory to forecast real-time data based on data collected. Numerous machine learning-
based techniques, including Support Vector Machine (SVM), autoregressive and moving average
(ARMA), Random Forest (RF), and back propagation neural network, are compared to the predicted
data. MEMS sensors do have complex mistakes, however the manufacturer may remove them
before the sensors are sold by employing calibration methods. In order to monitor the pertinent data,
four Wuhan stations have been located, and seven sensors have been installed. The seven sensors are
a rain-fall sensor, a humidity sensor, a pressure sensor, a pressure sensor for wind, a temperature
sensor, and a radiation sensor. The LSTM is seen to perform better than other machine learning
models.

3. Comparison of previous work done

Work done ML method used IoT device used Pros and Cons Accuracy/MAE
Arabelli et Logistic regression, DHT-11, BMP- + Real time data LR- 83.9%
al. 2023 [5] SVM, Decision tree, 180 + Execution time DT- 76%
 Random forest + Google colab is RF- 84.2%
used. SVM-83.6%
Shah et al. Decision tree regression, DHT-11, BMP- + Real time data 95% for Decision
2023 [12] random forest regression, 180 + Efficient tree
decision tree classifier, + Use of power 83% for Naïve
naïve Bayes classifiers Bi Bayes
Sathya, et Logistic Regression (LR) DHT11 + Effective LR- 92%
al.,2023 + Real time data
[14] - Work on
Limited data
Sharma, et Multiple linear microcontroller + Device is MLR- 81%
al., 2022 Regression, Logistic (Arduino wearable
[15] Regression UNO) , LM 35 + Expensive
Sensors
Xu et al., LSTM, PSO for NA + Optimized LSTM- 94%
2022 [16] optimization + Accuracy
- Time taking
Maliyeckel LightGBM, SVR, Hybrid None + Efficient For LightGBM
et al. 2021 model + Hybrid model MAE is 5.73
[17] is applied For SVR MAE is
+ Large scale 8.78
data is missing For Hybrid model
MAE is 4.90
Emmanuel Artificial Neural NA + Robust, With the lowest
et al., 2021 Networks (ANN) - Feed dependable, RMSE, MSE, and
[21] Forward Neural Network and reliable MAE of 6.360,
(FFNN), Cascade algorithms 40.45, and 0.54,
Forward Neural Network - Expensive respectively, the
(CFNN), Recurrent - Time- Elman NN has
Neural Network (RNN), consuming the greatest
and Elman Neural performance for
Network (ENN) the year 2018.
Shalini., et KNN, Decision tree, To collect real- + Effective Random forest:
al., 2021 Random forest, Linear time data from a + Good 0.083
[22] regression test region, an prediction Multiple linear
IoT-based - No specific regression: 0.165
weather station IoT device is Decision tree:
has been mentioned 0.094
developed. K-nearest
neighbour: 0.103
Syarifuddin A geometry algorithm on NA + The system 100% accuracy
et al 2021 an IoT platform to uses an IoT
[23] measure rainfall in real- platform,
time. which allows
for remote
monitoring and
data collection.
The time delay
Zhao et al LR, Support Vector Researchers - The model is QDA- 100%
2021 [24] Classification, Nearest used recordings effective in ET- 99%
Neighbour, Gaussian of steady rainfall predicting AdaBoost- 98.5%
Naïve Bayes, Bernoulli from five debris flow RF- 98.5%
Naive Bayes, AdaBoost, rainfall metres events based XGB – 98.3%
Gradient Boosting, in a debris flow on rainfall GB – 98%
Bagging, Random Forest, from December data. LR – 97.5%
Extra Trees, Gaussian 2012 to April - Requires BAGGING- 97%
Process Classification, 2015 hence no continuous KNN – 96.7%
Logistic Regression, IoT device is rainfall SVC – 96%
Extra Tree, Decision used. monitoring Nu-SVC -95.8%
Tree, Extra Tree, Gaussian NB -
Quadratic Discriminate 88%
Analysis.
Sadhukhan SVM, GPS-enabled - Accurate KNN- 70%
et al 2021 KNN IoT devices like - Portable LR- 84%
[25] DNN (Deep Neural HC-12, DHT- - Reliable SVM- 79.6%
Network) 11, BMP-180, - System may ANN- 89%
ANN (Artificial Neural BD-139 not able to
Network) (transistor) predict sudden
changes
Rani et al., Linear regression, Water float - Easy to adapt SVM- 90.60%
2020 [26] Logistic regression, sensors, Rain - Reliable ANN- 88%
Support vector machine, drop Sensors, - Some-time
ANN IoT Geck prediction may
not accurate
Onkar et al Rainfall prediction using Wireless Sensor - Flexible 100% Accuracy
2019 [27] machine learning Model Network (WSN) - Efficient
SVM GPRS (General detection of
Pocket Radio rainfall
Service) via a - Time delay
cellular network
Mzyece et Artificial Neural Network NA + Can handle ANN- 99%
al 2018 [28] (ANN) more data at
 one time Best validation
- Complex and using ANN- 91%
difficult to at epoch 44
interpret.
Salmayenti Artificial Neural Network NA + High accuracy High prediction
et al (ANN) for rainfall - Accuracy accuracy for
2017[29] prediction decreases monsoonal
when used for regions (R2:
long-term 0.59-0.82,
predictions. RMSE: 0.04-
0.09)

4. Visualization of previous work

This section of the chapter focuses on the comparative study of previous work done so that reader
can have good idea about the work and percentage of accuracy achieved in this field. In order to
present this section some previous standard work has been compared in the consequent section of
the chapter.

Islam M. S. 2023 [30] have done a study for rain prediction and for that various machine learning
algorithms like Radial bias classifier, Multilayer perceptron, K nearest neighbour, SVM, Decision
tree, Logistic regression, CatBoost, LightGBM, Gaussian Naïve bayes, AdaBoost, GRU, LSTM,
CNN have been applied and it is found that CNN with 100 epoch perform better in terms of
accuracy. Figure 1 shows the comparison of the machine learning models in terms of accuracy,
RMSE, MSE, MAE etc. Islam M. S. 2023 [30] have outlined several strategies for enhancing the
performance for rain fall prediction using AI and IoT. By leveraging XGBoost's ensemble learning
abilities, they seek to enhance the model's precision in forecasting rainfall events. In this study
CatBoost and LightGBM, two ensemble models are compared and, both models consistently
demonstrate higher accuracy compared to other models. This might indicate the strength of
ensemble techniques in handling the complexities of the rainfall prediction task.
 100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%

Accuracy RMSE MSE MAE

Figure1: Performance comparison of Prediction Model [30]

According to research by Islam et al.[31], Linear Gradient Boosting (LGB) with Select-K-Best
feature selection exhibits the best performance in terms of regression, with test set scores of 0.203,
MAE of 6.40, and RMSE of 15.44. Among the classifiers, XG-Boost (XGB) with no feature
selection and no sampling has the greatest accuracy of 0.787 and the f1-score of 0.62 on the test set.
With an R2-score of 0.189, an MAE of 5.789, and an RMSE of 15.575, the XGB classifier paired
with LGB regression without any feature selection performed the best on the test set among the ZIR
models. The ZIR models provide lower MAE scores and somewhat higher RMSE scores than
regression models, despite being outperformed by them in terms of r2-scores. Figure 2 compares
both the traditional LGB models and the XGB classifier/LGB regression combination.

20
15.44 15.575
15
R2-score

10
6.4 5.789 MAE
5
0.203 0.189
0 RMSE
LGB(in Regression) XGBC,LGBR

Figure 2: Comparison of LGB, and combined XGBC, LGBR [31]

Lagrazon et al. 2023 [32] have provided a comparison analysis of machine learning models for
performance evaluation of several machine learning models like Tree, SVM, Gaussian Process
Regression, Ensemble, and Neural Network models. With the lowest RMSE and greatest R-squared
values within this group of models, the Gaussian Process Regression model stood out as having the
best overall performance. The potential to improve flood preparedness and early warning systems
makes these discoveries significant. It is possible to create a programme that can accurately
anticipate rainfall levels by integrating a well-validated machine learning algorithm. Comparing the
effectiveness of various models is shown in Figure 3.

1.2
1
0.8
0.6
0.4
0.2
0
Support Gaussian
Neural
Tree Vector Process Ensemble
Network
Machine Regression
RMSE 0.000812 0.00089 0.000362 0.00061 0.001173
Rsquared 0.999277 0.999132 0.999856 0.999592 0.998492
MAE 0.000446 0.000735 0.000208 0.000317 0.000818

RMSE Rsquared MAE

Figure 3: Performance comparison for different models [32]

Maliyeckel et al. 2021 [17] have introduced a novel hybrid model that synergizes LightGBM and
SVR algorithms through ensemble modelling, resulting in more accurate precipitation predictions
for specific topographical locations. This innovative approach, utilizing a weighted average of
algorithmic outputs, outperformed traditional standalone SVR and LightGBM models. Notably, the
hybrid model exhibited the lowest root mean square error, underscoring its superior predictive
capability. Despite encountering constraints such as limited sample size and data accessibility, the
study harnessed weather data from diverse stations to validate the hybrid model's efficacy. Figure 4
shows the performance comparison between LightGBM, SVR and Hybrid model in terms of AME
and RMSE [17] and it is very clear that hybrid based method shows the good performance as
compare to other methods.
 8
6.86
7 6.2
5.73 5.79
6
4.9
5
3.82
4 MAE
3 RMSE
2
1
0
LGBM SVR Hybrid

Figure 4: Performance comparison for LightGBM, SVR and Hybrid method [17]

120

97.5 98.5 96.7

100 93
88
83.9 84.2 83.6 83 84
79.6
80 76
LR
70
DT
60 RF
NB
40
SVM
KNN
20

NA NA NA NA NA NA NA NANANA
0
Arabelli et al. 2023 Shah et al. 2023 [12] Zhao et al 2021[24] Sadhukhan et al 2021
[5] [25]

Figure 5: Performance comparison of various models [5], [12], [24], [25]

Figure 5 shows the performance comparison of various ML models for the year 2023 and 2021
articles and shows that the highest prediction accuracy is achieved for the Random forest method,
then for Linear regression, then for KNN etc. The prediction accuracy is also highly dependent upon
the data availability for training and testing, amount and quality of data related to rain prediction.

Conclusion

Rain prediction is one of the important concerns for everyone especially for the farmers and with
new technology like IoT and AI, now it is possible to predict rain more precisely and accurately.
Rain prediction is more dependent on three factors like humidity, temperature and wind speed. It has
also noticed that rain prediction using IoT based devices, increases the accuracy by 2-3 percentage.
It is also important to consider that amount of data used for prediction is also an impact factor and
good types of IoT device may also lead to good model. Sometime hybrid based model may provide
good accuracy as compare to other traditional model. So before developing any rain-prediction
model all the factors and resources available should keep into accounts for good model.

Abbreviation used

Abbreviation Full word

AI Artificial intelligence
IoT Internet of Things
MLR Multivariate linear regression
RF Random Forest
XG-Boost Extreme gradient boost
LR Logistic Regression
ELSWP Enhanced Learning Scheme for Weather Prediction
MAE Mean absolute error
RMSE Root mean square error
SD Standard deviation
KNN K-nearest neighbour
ANN Artificial Neural Network
ELM Extreme learning machine
WKNN Weighted K nearest neighbour
DWKNN Dual weighted K nearest neighbour
MEMS Micro-electromechanical systems
ARMA Autoregressive and moving average
SVM Support vector machine
FFNN Feed Forward Neural Network
CFNN Cascade Forward Neural Network
RNN Recurrent Neural Network
ENN Elman Neural Network
DNN Deep Neural Network

References

1. Gupta, A., Mall, H. K., & Janarthanan, S. (2022). Rainfall Prediction Using Machine
Learning. In 2022 First International Conference on Artificial Intelligence Trends and
Pattern Recognition (ICAITPR) (pp. 1-5). Hyderabad, India. doi:
10.1109/ICAITPR51569.2022.9844203.

2. Costa, V. G., & Pedreira, C. E. (2023). Recent advances in decision trees: An updated
survey. Artificial Intelligence Review, 56(5), 4765-4800.
3. Jaiswal, S., Pandey, M. K. (2023). Digital Watermark Extraction Using RS-KNN and RS-
LDA with LWT and Statistical Features. SN COMPUT. SCI. 4, 496.
https://doi.org/10.1007/s42979-023-01924-9
4. Jaiswal, S., Pandey, M.K. (2023). Deep Artificial Neural Network Based Blind Color Image
Watermarking. In: Bhattacharyya, S., Banerjee, J.S., De, D., Mahmud, M. (eds) Intelligent
Human Centered Computing. Human 2023. Springer Tracts in Human-Centered Computing.
Springer, Singapore. https://doi.org/10.1007/978-981-99-3478-2_10.
5. Arabelli, R., Sindhu, C., Keerthana, M., Madhav, T.V., Vamshi, C., Ahmed, S.M. (2023).
Improved Numerical Weather Prediction Using IoT and Machine Learning. In: Kumar, A.,
Mozar, S., Haase, J. (eds) Advances in Cognitive Science and Communications. ICCCE
2023. Cognitive Science and Technology. Springer, Singapore. https://doi.org/10.1007/978-
981-19-8086-2_109.
6. Rehman, A., Tariq, S., Farrakh, A., Ahmad, M., &Javeid, M. S. (2023). A Systematic
Review of Machine Learning and Artificial Intelligence Methods to Tackle Climate Change
Impacts. In 2023 International Conference on Business Analytics for Technology and
Security (ICBATS) (pp. 1-7). Dubai, United Arab Emirates. doi:
10.1109/ICBATS57792.2023.10111347.
7. Singh, S. K., Tiwari, A. K., & Paliwal, H. K. (2023). A state-of-the-art review on the
utilization of machine learning in nanofluids, solar energy generation, and the prognosis of
solar power. Engineering Analysis with Boundary Elements, 155, 62-86.
8. Praveen, B., Talukdar, S., Shahfahad et al. (2020). Analyzing trend and forecasting of
rainfall changes in India using non-parametrical and machine learning approaches. Sci
Rep 10, 10342. https://doi.org/10.1038/s41598-020-67228-7.
9. Nandgude, N., Singh, T. P., Nandgude, S., & Tiwari, M. (2023). Drought Prediction: A
Comprehensive Review of Different Drought Prediction Models and Adopted
Technologies. Sustainability, 15(15), 11684.
10. Dowlut, N., & Gobin-Rahimbux, B. (2023). Forecasting resort hotel tourism demand using
deep learning techniques–A systematic literature review. Heliyon.
11. Molina, M. J., O’Brien, T. A., Anderson, G., Ashfaq, M., Bennett, K. E., Collins, W. D., ...
& Ullrich, P. A. (2023). A Review of Recent and Emerging Machine Learning Applications
for Climate Variability and Weather Phenomena. Artificial Intelligence for the Earth
Systems, 1-46.
12. P. Shah, Y. A, U. Khaitan and S. Kayalvizhi, "Weather Management System Using Machine
Learning Algorithm And IOT," 2023 International Conference on Recent Advances in
Electrical, Electronics, Ubiquitous Communication, and Computational Intelligence
(RAEEUCCI), Chennai, India, 2023, pp. 1-4, doi:
10.1109/RAEEUCCI57140.2023.10134454.
13. Liyew, C.M., Melese, H. A. (2021). Machine learning techniques to predict daily rainfall
amount. J Big Data 8, 153. https://doi.org/10.1186/s40537-021-00545-4.
14. Sathya, P., Gnanasekaran, P. (2023). Rainfall Forecasting System Using Machine Learning
Technique and IoT Technology for a Localized Region. In: Shakya, S., Du, KL., Ntalianis,
K. (eds) Sentiment Analysis and Deep Learning. Advances in Intelligent Systems and
Computing, vol. 1432. Springer, Singapore. https://doi.org/10.1007/978-981-19-5443-6_32.
15. Sharma, S.A., Gangopadhyay, A., Koushik, K.T., Sriharipriya, K.C., Clement, J.C. (2022).
Real-Time Rainfall Prediction System Using IoT and Machine Learning. In: Arunachalam,
V., Sivasankaran, K. (eds) Microelectronic Devices, Circuits and Systems. ICMDCS 2022.
Communications in Computer and Information Science, vol 1743. Springer, Cham.
https://doi.org/10.1007/978-3-031-23973-1_10
16. Xu, Y., Hu, C., Wu, Q., Jian, S., Li, Z., Chen, Y., Zhang, G., Zhang, Z., & Wang, S. (2022).
Research on particle swarm optimization in LSTM neural networks for rainfall-runoff
simulation. Journal of Hydrology, 608, 127553.
https://doi.org/10.1016/j.jhydrol.2022.127553
17. Maliyeckel, M. B., Sai, B. C., & Naveen, J. (2021). A Comparative Study of LGBM-SVR
Hybrid Machine Learning Model for Rainfall Prediction. In 2021 12th International
Conference on Computing Communication and Networking Technologies (ICCCNT) (pp. 1-
7). Kharagpur, India. doi: 10.1109/ICCCNT51525.2021.9579628.
18. Dash, Y., Mishra, S. K. Panigrahi, B. K. (2018). Rainfall prediction for the Kerala state of
India using artificial intelligence approaches. Computers and Electrical Engineering, 70.
https://doi.org/10.1016/j.compeleceng.2018.06.004.
19. Huang, M., Lin, R., Huang, S., & Xing, T. (2017). A novel approach for precipitation
forecast via improved K-nearest neighbor algorithm. Advances in Engineering Informatics,
33, 89-95. doi: 10.1016/j.aei.2017.05.003.
20. Chao, Z., Pu, F., Yin, Y., Han, B., & Chen, X. (2018). Research on Real-Time Local Rainfall
Prediction Based on MEMS Sensors. Journal of Sensors, 2018, e6184713. doi:
10.1155/2018/6184713.
21. Emmanuel, Gbenga, Dada., Hurcha, Joseph, Yakubu., David, Opeoluwa, Oyewola. (2021).
Artificial Neural Network Models for Rainfall Prediction. doi:
10.24018/EJECE.2021.5.2.313
22. Shalini, H., Aravinda, C.V. (2021). An IoT-Based Predictive Analytics for Estimation of
Rainfall for Irrigation. In: Chiplunkar, N.N., Fukao, T. (eds) Advances in Artificial
Intelligence and Data Engineering. AIDE 2019. Advances in Intelligent Systems and
Computing, vol 1133. Springer, Singapore. https://doi.org/10.1007/978-981-15-3514-7_105
23. Syarifuddin, S. D. S., Khurniawan, A., Aulia, S., Ramadan, D. N., & Hadiyoso, S. (2021).
Rainfall Information System Using Geometry Algorithm on IoT Platform. In 2021 IEEE
Asia Pacific Conference on Wireless and Mobile (APWiMob) (pp. 195-199). Bandung,
Indonesia. doi: 10.1109/APWiMob51111.2021.9435219.
24. Yan, Zhao., Xingmin, Meng., Tianjun, Qi., Yajun, Li., Guan, Chen., Dongxia, Yue., Feng,
Qing. (2021). AI-based rainfall prediction model for debris flows. Engineering Geology,
106456-. doi: 10.1016/J.ENGGEO.2021.106456
25. Sadhukhan, M., Dasgupta, S.and Bhattacharya, I. (2021). An Intelligent Weather Prediction
System Based on IOT 2021 Devices for Integrated Circuit (DevIC), Kalyani, India, pp. 528-
532, doi: 10.1109/DevIC50843.2021.9455883.
26. Rani, D. S., Jayalakshmi, G. N., & Baligar, V. P. (2020). Low Cost IoT based Flood
Monitoring System Using Machine Learning and Neural Networks: Flood Alerting and
Rainfall Prediction. 2020 2nd International Conference on Innovative Mechanisms for
Industry Applications (ICIMIA). doi:10.1109/icimia48430.2020.9074.
27. Onkar, Amale., Rupali, Patil. (2019). IOT Based Rainfall Monitoring System Using WSN
Enabled Architecture. doi: 10.1109/ICCMC.2019.8819721.
28. Mzyece, L., Nyirenda, M., Kabemba, M. K., & Chibawe, G. (2018). Forecasting Seasonal
Rainfall in Zambia – An Artificial Neural Network Approach. Zambia ICT Journal, 2(1),
16–24. https://doi.org/10.33260/zictjournal.v2i1.46
29. Salmayenti, R., Hidayat, R. & Pramudia, A. (2017). Rainfall Prediction Using Artificial
Neural Network. Jurnal Agromet Indonesia, 31(1), 11-21.
https://doi.org/10.29244/J.AGROMET.31.1.11-21
30. Islam, M. S., Hossain, A., Khatun, A., Kor, A. (2023). Evaluation of the Performance of
Machine Learning and Deep Learning Techniques for Predicting Rainfall: An Illustrative
Case Study from Australia. DOI: 10.1109/ECCE57851.2023.10101560.
 31. Islam, F. Z., Islam, R., Jamil, A., & Momen, S. (2022). Evaluation of Machine Learning
Methods for Predicting Rainfall in Bangladesh. DOI: 10.1109/CITDS54976.2022.9914171.
32. Lagrazon, P. G. G. et al., (2023). A Comparative Analysis of the Machine Learning Model
for Rainfall Prediction in Cavite Province, Philippines. IEEE World AI IoT Congress
(AIIoT), Seattle, WA, USA, 2023, pp. 0421-0426, doi: 10.1109/AIIoT58121.2023.10174533.

View publication stats