IRJET- Crop Prediction System using Machine Learning Algorithms

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 07 Issue: 02 | Feb 2020 www.irjet.net p-ISSN: 2395-0072
© 2020, IRJET | Impact Factor value: 7.34 | ISO 9001:2008 Certified Journal | Page 748
Crop Prediction System using Machine Learning Algorithms
Pavan Patil1, Virendra Panpatil2, Prof. Shrikant Kokate3
1Pavan Patil SPPU, (Pimpri Chinchwad College of Engineering)
2Virendra Panpatil SPPU, (Pimpri Chinchwad College of Engineering)
3Professor Shrikant Kokate, Dept. of Computer Engineering, PCCOE, Maharastra, India
---------------------------------------------------------------------***----------------------------------------------------------------------
Abstract - As we are aware of the fact that, most of Indians
have agriculture as their occupation. Farmersusuallyhave the
mindset of planting the same crop, using more fertilizers and
following the public choice. By looking at the past few years,
there have been significant developments in how machine
learning can be used in various industries and research. So we
have planned to create a system where machine learning can
be used in agriculture for the betterment of farmers. The
surveyed research papers have given a rough ideaaboutusing
ML with only one attribute. We have the aim of adding more
attributes to our system and ameliorate the results, whichcan
improve the yields and we can recognize several patterns for
predictions. This system will be useful to justifywhichcropcan
be grown in a particular region
Key Words: Machine Learning in Agriculture,
Classification Algorithms, Decision Tree, KNN
1. INTRODUCTION
Crop production may be a complicated development that's
influenced by soil and environmental condition input
parameters. Agriculture input parameters vary from field to
field and farmer to farmer. Collection such info on a bigger
space may be a discouraging task. However, the
environmental condition info collected in Republic of India
at each 1sq.m space in numerous components of the district
is tabulated by Indian meteoric Department. The massive
such knowledge sets may be used for predicting their
influence on major crops of that individual district or place.
There are completely different foretelling methodologies
developed and evaluated by the researcherseverywherethe
globe within the field of agriculture or associated sciences. A
number of such studies are: Agricultural researchers in
alternative countries have shown that tries of crop yield
maximization through pro-pesticide state policies have LED
to hazardously high chemical usage. These studies have
reported a correlation between chemical usage and crop
yield [1]. Agriculture is associate trade sector that's
benefiting powerfully from the event of detectortechnology,
knowledge science, and machine learning (ML) techniques
within the latest years. These developmentsreturntosatisfy
environmental and populationpressuresround-facedby our
society, wherever reports indicate a requirement for robust
international agriculture yield increasetoproducefoodfora
growing population on a hotter planet. Most of the work
tired the sector of yield foretelling via cubic centimeter
makes use of some kind of remote sensing knowledge over
the farm. Agriculture seeks to extend and improve the crop
yield and therefore the quality of the crops to sustainhuman
life. However, within the current time, folkstendto requirea
lot of like a shot appreciated jobs.Therearefewer,andfewer
folks concerned in crop cultivation. additionally, the
continual increase of human population makes the
cultivation of the crops at the proper time and right place
even a lot of vital, because the climate is dynamic and
therefore the shifts fromtraditional weatherpatternarea lot
of frequent than before manufacture.Foodinsecuritymaybe
a drawback that can't be avoided, and humans should build
use of latest innovative technologies to create useofexisting
soil, water and air conditions to get larger crops. The
information gap between ancient ways that of cultivating
and new agricultural technologies may be overcome if the
computer code may be designed to model the interactive
impact of climate factors, particularly the impact of
maximum events (e.g. heat, rainfalls and excess water)
occurring at completely different growing phases of crops.
The temperature change undoubtedly affects the native and
world food production, therefore planningcomputercode to
model crop predictions needs new methodology for
temperature change studies, situations for temperature
change adaptation, and policymakers which will limit the
devastating effectsofweatheronfoodprovide.Experimental
proof is employed to form environmental condition zones
that have seen changes in weather and water, the 2 most
significant factors in guaranteeing a incrop.Thesoil sortwill
modification over time because of weather and pests,
therefore crop managementmustmanagea fancyquantityof
information, directly or indirectly associated with one
another. It will therefore by considering a simplified reality,
to permit a quick assessment of the impact of temperature
change in agriculture. Agriculture should adapt to those
climate changes, and it will do therefore by developing
models which will in theoryoptimizemanagementpractices,
maximize the rotations of the new crop to manage the
changes of soil, novel breeding programs.By maximizing the
worth of foretelling, the seasonal climate changes may be
ascertained and recorded in an exceedingly timely manner.
Later on, by victimizationcomputercodesupportedmachine
learning, one will timely assess the temperature change
impact and check attainable situations that incorporate
ascertained changes in climatic conditions and water
distribution. data {processing} is that the process of
analyzing the experimental knowledge collected over a
amount and varied locations from completely different
views, extract trends or patterns {of data of knowledge of
info} and switch them into helpful information for users.
Users will then additionally reason and/or summarize the
relationships ascertained from the collectedknowledge,and

typically predict what knowledge to expect. Machine-
learning techniques are a part of data processing and
knowledge exploration and focus exclusively on
characteristic correlations or patterns among massive
datasets or massive relative databases. The patterns,
associations, or relationships among all this knowledge will
additional be reborn into information that's offered to the
user as historical patterns and future trends. This
information provided by machine learning will facilitate
farmers with crop cultivation by predicting probabilities of
crop losses or stop losses altogether.
2. Literature Review
Nowadays many experts are applying automated farming.
Since Decision Tree is an well-known algorithm it was used
for prediction which is a supervised learning algorithm and
multiple linear regression which is generalized prediction
model. An attempt has been made to research the influence
made by decision tree induction technique of climatic
parameters on soybean productivity.Foreasyunderstanding
of end-user different kind of rules were created from the
Decision tree. The paper from Md. Tahmid Shakoor & co
paper helped us for selecting various attributes like land
capability classification, soil depth, slope, drainage, texture,
erosion, and permeability [4]. Two supervised classification
machine learning algorithms has been implemented in this
study. Our system takes the necessary weather and soil
properties data for a given coordinate automatically from an
appropriate source. Another advantage is that their system
worked on large regions, and provides forecasts at a
resolution compatible with best input data resolution, which
in the case is originally from the soil data. The ability of
forecasting crop before the beginningofthecropseason.This
provides users with the capability to perform strategy
changes, like choosing a more robustgeneticvariationbefore
planting or even changing the crop type, in order to
accommodate for extreme climatic variations further ahead
in the crop cycle [2].
The algorithm developed introduces a data-driven model to
predict and forecast crop yield using joint dependencies of
soil and climate features. Although there are several
techniques existing to obtain rainfall predictions, the
algorithm discussed in this paper succeeded in emphasizing
on Rainfall along with the crop yield prediction. This
designed model took into account the most relevant
environment as well as soil parameters that affect the crop
growth, in a way that each of those parameters received
equal weight in the final prediction. The outcomes of this
research can benefit the agriculturists/farmers by knowing
the investment capital on the crop to be sown, even before
the sowing season begins. The predictive pattern of the
algorithm can benefit local self-government and financial
institutions to allocate suitable funds or fiscal loans to
farmers. Naive Bayes is used for the large dataset can also be
beneficial. Use of naïve Bayes and decision tree makes the
model very efficient in terms of computation. The system is
scalable as it can be used to test on different crops. From the
yield graphs, the best time of sowing, plant growth and
harvesting of the plant canbefoundout.Also,theoptimaland
worst environmental condition can also be incurred. The
model focuses on all type of farms, and smaller farmers can
also be benefitted. Thismodelcanbefurtherenhancedtofind
the yield of every crop, and for pesticide recommendation.
Also, it can be modified to suggest about the fertilizers and
irrigation need for crops.
3. Related Work
1. A Scalable Machine Learning System for Pre-Season
Agriculture Yield Forecast:
The system projected during this work is created by a neural
network wherever inputs area unit treated on an individual
basis. Static soil information in handled by fully-connected
layers whereas dynamic meteorological information is
handled by continual LSTM layers. This explicit design was
trained with historical information for many soil properties,
precipitation, minimum and most temperature against
historical yield labels at county level. When training, the
model was tested in an exceedingly separate information set
and showed comparable results with existing yield
prognostication ways that create use of in-depth remote
sensing data. the most important lesson learnt from our
experiments is that it's attainable get ascendable yield
forecast as a result of the projected neural network model
will notice and exploit redundant info each within the soil
and within the weather information. To boot, the model
might be able to learn AN implicit illustration of the cycles of
the crops evaluated during this paper, considering the
seasonal atmospherically information used as input.
2. Machine learning approach for forecasting crop yield
based on climatic parameters
The present study provides the potential use of information
mining techniques in predicting the crop yieldsupportedthe
environmental condition input parameters. The developed
webpage is user friendly and therefore the accuracy of
predictions squaremeasurehigherthanseventy-fivepercent
all told the crops and districts designated within the study
indicating higher accuracy of prediction. The user-friendly
web content developed for predicting crop yield may be
utilized by any user their alternative of crop by providing
environmental condition knowledge of that place.
3. Crop Prediction on the RegionBeltsofIndia:ANaïveBayes
MapReduce Precision Agricultural Model
The planned work introduces efficient degree economical
crop recommendation system. Use of naïve mathematician
makes the model terribly economical in terms of
computation. The system is scalable because it may be wont
to take a look at on totally different crops. From the yield
graphs the simplest time of sowing, plant growth and gather
of plant may be known. Conjointly the best and worst
condition may also be incurred. The model focuses on all
style of farms, and smaller farmers may also be benefitted.

This model may be more increased to seek out the yield of
each crop, and for chemical recommendation. Conjointly it
may be changed to recommend concerningthefertilizersand
irrigation want of crops.
4. Evaluation of Predictive Data Mining Algorithms in Soil
Data Classification for Optimized Crop Recommendation.
In this study, we've given the analysis potentialities for the
classification of soil by mistreatment well-known
classification algorithms as J48, BF Tree, and OneRandNaïve
Bayes; in data processing. The experiment was conductedon
information instances from Kasur district, Pakistan.Wehave
ascertained the comparative analysis of those algorithms
have the various level of accuracy to determine the
effectiveness and potency of predictions. However, the
advantages of the higher understanding of soils classes will
improve the productivity in farming, reduce dependence on
fertilizers and build higher prognostic rules for the advice of
the rise in yield. In the future,we have a tendency to contrive
to form a Soil Management and
5. Agricultural Production Output Prediction Using
Supervised Machine Learning Techniques
Two supervised classification machine learning formula has
been enforced during this study. the choice Tree Learning-
ID3 (Iterative Dichotomiser 3) and KNNR discover the
patterns within the knowledge set containing average
temperature and precipitation worth obtained throughout
the cropping amount of six major crops in 10 major cities of
Bangladesh for the past twelve years and provides the
prediction. ID3 uses the choice tree table that consists of the
rangesoftheprecipitation,temperatureandyieldknowledge.
The research provides an answer to the current downside
that was much required for farmers in People's Republic of
Bangladesh. Though the research is restricted to some
mounted dataset, the long run ahead promises addition of a
lot of knowledge which will be analyzed with more machine
learning techniques to come up with crop predictions with
higher exactness. Moreover, the analysis will result in profits
and invention of advanced farming techniques which will
improve our economy and can facilitate United States stand
out as a technologically advanced country.
4. EXISTING SYSTEM
An agro-based country depends on agriculture for its
economic growth. When a population of the country
increases dependency on agriculture also increases and
subsequent economic growth of the country is affected. In
this situation, the crop yield rateplays asignificantroleinthe
economic growth of the country. So, there is a need to
increase crop yield rate. Some biological approaches (e.g.
seed quality of the crop, crop hybridization, strong
pesticides) and some chemical approaches (e.g. use of
fertilizer, urea, potash) are carried out to solve this issue. In
addition to these approaches,a crop sequencing techniqueis
required to improve the net yield rate of the crop over the
season. One of existing system weidentified isCropSelection
Method (CSM) to achieve a net yield rate of crops over the
season. We have taken exampleofCSMtodemonstratehowit
helps farmers in achieving more yield
Crop can be classified as:
a) Seasonal crops— crops can be plantedduringaseason.e.g.
wheat, cotton.
b) Whole year crops— cropscan be planted duringtheentire
year. e.g. vegetable, paddy, Toor.
c) Short time plantation crops— crops that take a short time
for growing. e.g. potato, vegetables, ratio. d) Long-time
plantation crops— These crops take a long time for growing.
e.g. sugarcane, Onion. A combination of these crops can be
selected in a sequence based on yield rateper day. Illustrates
sequences of crops with cumulative yield rate over the
season. CSM method, shown in may improve the net yield
rate of crops using the limited land resource and also
increases re-usability of the land.
Basically, in crop selection method makes use of technique
where it recommends different set of crops for same area
over the years. There are various options are available to
select for farmers. They can choose one of the options and
observe the results. The combination which will give high
yield for same area is generated as output for that area. In
this way CSM method tries to predict the suitable crops for
given area.FarmingSystemsinIndiaarestrategicallyutilized,
according to the locations where they are most suitable. The
agricultural systems that significantly follows to the
agricultureof Indiaare subsistencefarming,organicfarming,
industrial farming. Regions all over India differ in types of
farming they use; some are based on horticulture, ley
farming, agroforestry, and many more. The surveyed
research papers have given a rough ideaaboutusingMLwith
only one attribute. We havetheaimofaddingmoreattributes
to our system and amelioratethe results, which can improve
the yields and we can recognize several patterns for
predictions. This system will be useful to justify which crop
can be grown in a particular region

5. PROPOSED SYSTEM
In our system we are making use of a classification algorithms to improvised the crop yields.
5.1. Data Acquisition:
Dataset must have following attributes
 Soil Parameters:
Soil Type
Soil Ph value
 Climatic Parameters:
Humidity
Temperature
Wind
Rainfall
 Production
Cost of cultivation
Previous year yield details for that region
In this project we are performing crops prediction for
district level. So main aim is to find the dataset which
contains production details of past 10-12 years also details
about climatic parameters and soil parameters like rainfall,
temperature, moisture, soil contents etc. details. These
factors will help in the prediction of the crops by using
various classifiers on the given dataset. Thus,variousfactors
are assessed and the factors strongly leading to accurate
prediction of the crops
5.2 Preprocessing:
The dataset that is used needs to be pre-processed because
of the presence of redundant attributes, noisy data in it.
Initially, data cleaning operation is performed where the
redundant factors aredeterminedandarenotconsideredfor
the prediction of crops. Over18 which are either having the
same values for all the employees or are completely
unrelated to the prediction task. As part of the exploratory
data analysis, the categorical factors are split and are
assigned values as 0 and 1 based on whether the factor is
present or not. These assigned values assist in further
classification based on that particular factor.
5.3 Classifier Models:
5.3.1 Decision Tree Classifier:
The decision tree is method of selectingbestroot nodesuntil
we get elements of same class we keep on splitting the tree
on the basis of attributes. With versatile features helping

actualize both categorical and continuous dependent
variables, it is a type of supervisedlearningalgorithmmostly
used for classification problems.Whatthisalgorithmdoes is,
it splits the population into two or more homogeneous sets
based on the most significant attributes making the groups
as distinct as possible. The decision tree algorithm will give
us best split on different features for selection of most
suitable crop among the population. The feature selection
methodology of Decision tree classifier makes it suitable for
prediction of suitable crops. The Selection attributes of
Decision tree classifier are as follow.
5.3.1.1 Gini Index
Gini index says, if we select two items from a population at
random then they must be of same class and probability for
this is 1 if population is pure. Used to calculate impurity for
the features of given classes.
5.3.1.2 Entropy
A decision tree is built top-down from a root node and
involves partitioning the data into subsets that contain
instances with similar values (homogeneous). If the sample
is completely homogeneous the entropy is zero and if the
sample is equally divided then it has entropy of one.
5.3.1.3 Information Gain
The information gain is based on the decrease in entropy
after a dataset is split on an attribute. Constructing a
decision tree is all about finding attribute that returns the
highest information gain (i.e., the most homogeneous
branches. This attributeselectionmethodswill playvital role
in prediction of crop.
5.3.1.4 C4.5 Algorithm
The C4.5 algorithmic program uses info gain as ripping
criteria. It will handle numerical and categorical information
similarly as missing values. To handle continuous values, it
generates threshold and so divides attributes with prices
quite the edge price and values up to the edge value. It offers
the subsequent edges. They’re explicable, in contrast to
different classifiers, that need to be seen as a recorder that
has a class to a given input instance. Call trees will be
envisioned as tree graphs wherever nodes and branches
represent the classification rules learnt, and leaves denote
the ultimate categorizations.
5.3.3 KNN
KNN may be a variety of instance-based learning, wherever
the performance is barely approximated regionally and
every one computation is delayed till it's the classification.
Both for classification and regression, a helpful technique
will be to assign weights to the contributions of the
neighbors, in order that the nearer neighbors contribute
additional to the typical than the additional distant ones.
6. RESULTS AND ANALYSIS
We tested decision tree, naïve bayes classifier, and KNN
classifier with sample dataset containing attributeslike crop
name, cost of cultivation, costofirrigation,costofproduction
which are independent variables and yield per hectare is
dependent variable.
The result obtained are represented using confusion matrix
which shows relation between prediction of algorithms and
actual values obtained when sample are testedaftertraining
dataset.
1. Decision Tree Classifier:
Confusion matrix for decision tree classifier:
[ [221, 49],
[67, 163]]
Accuracy for Decision-Tree: 76.8%
Precision for Decision-Tree:0.767
Specificity for Decision-Tree:0.708
2. KNN Classifier:
Confusion matrix for KNN:
[[269, 30],
[23, 178]]
Accuracy for KNN: 89.4%
Precision for KNN:0.921
Specificity for KNN:0.8825
Accuracy Comparison:

Precision Comparison
Specificity Comparison:
7. CONCLUSION
The project work introduces an efficient crop
recommendation system using classifiermodels.Thesystem
is scalable as it can be used to test on different crops. From
the yield graphs the best time of sowing, plant growth and
harvesting of plant can also be found out along with
prediction for crops. Decision tree shows poor performance
when dataset is having more variations but naïve bayes
provides better result than decision tree for such datasets.
The combination classification algorithm like naïve bayes
and decision tree classifier are better performingthanuse of
single classifier model.
REFERENCES
[1]. S.Veenadhari, Dr Bharat Misra, Dr CD
Singh.2019.”Machine learning approachforforecastingcrop
yield based on climatic parameters.”.978-1-4799-2352-
6/14/$31.00 ©2014 IEEE
[2]. Igor Oliveira, Renato L. F. Cunha, Bruno Silva, MarcoA.S.
Netto.2018.”A Scalable Machine Learning System for Pre-
Season Agriculture Yield Forecast.”.978-1-5386-9156-
4/18/$31.00 ©2018 IEEE DOI
10.1109/eScience.2018.00131
[3]. Neha Rale, Raxitkumar Solanki, Doina Bein, James
Andro-Vasko, Wolfgang Bein.”Prediction of Crop
Cultivation”.978-1-7281-0554-3/19/$31.00©2019 IEEE
[4]. Md. Tahmid Shakoor,Karishma Rahman,Sumaiya Nasrin
Rayta, Amitabha Chakrabarty.2017.”Agricultural Production
Output Prediction Using Supervised Machine Learning
Techniques”.978-1-5386-3831-6/17/$31.00 ©2017 IEEE
[5]. G Srivatsa Sharma, Shah Nawaz Mandal, Shruti Kulkarni,
Monica R Mundada, Meeradevi.2018.”Predictive Analysis to
Improve Crop Yield Using a Neural Network Model”.978-1-
5386-5314-2/18/$31.00 ©2018 IEEE
[6]. Rashmi Priya, Dharavath Ramesh.2018.”CropPrediction
on the Region Belts of India: A Naïve Bayes MapReduce
Precision Agricultural Model”. 978-1-5386-5314-
2/18/$31.00 ©2018 IEEE
[7]Talha Siddique,Dipro Barus,Zanntual Fredous,Amitabh
Chakravarti. 2017. “AutomatedFarmingPrediction”.0978-1-
5090-6182-2/17/$31 @2017 IEEE
[8]Takeshi Yoshida Noriyuki Murakami and Hiroyuki
Tauiji.2017. Hybrid Machine Learning Approach to
Automatic Plant PhenotypingForSmartAgriculture”. 978-1-
5090-5888-4/16/$31.00 @IEEE 2016
BIOGRAPHIES
Prof. Shrikant Kokate
Qualification: ME Comp
Area of Interest: Data Science,Web
Technologies
Name: Pavan Patil
Qualification:BE
Computer(Pursuing)
Name: Virendra Panpatil
Qualification:BE
Computer(Pursuing)
2nd
Author
Photo
3rd
Author
Photo

IRJET- Crop Prediction System using Machine Learning Algorithms

Recommended

More Related Content

What's hot (20)

Similar to IRJET- Crop Prediction System using Machine Learning Algorithms (20)

More from IRJET Journal (20)

Recently uploaded (20)

IRJET- Crop Prediction System using Machine Learning Algorithms