A Mobile Application Through Crop Yield Is Predicted Using Data Mining Techniques

Download as pdf or txt
Download as pdf or txt
You are on page 1of 4

Volume 3, Issue 4, April – 2018 International Journal of Innovative Science and Research Technology

ISSN No:-2456-2165

A Mobile Application Through Crop Yield is Predicted


using Data Mining Techniques
Thanushree C Harshitha .P Manjunath C R
Department of Computer Science and Department of Computer Science and Department of Computer Science and
Engineering Engineering Engineering Associate professor
Jain University Jain University Jain University
Bangalore, India Bangalore, India Bangalore, India

Abstract:- The economy of India is mainly dependent on dependent on agriculture. Agriculture also play an important
agriculture. Most of the people in India take up role in the development of socio-economic fabrication of India
agriculture as their occupation. India also suffers from and which is demographically a broad sector of income.
natural calamities like flood, drought are the main reasons Agriculture supports the livelihood of about 58 per cent of
for the crop damage and huge loss to the farmers. So the India’s population. Agriculture is the major contributor to
crop yield prediction in advance to harvest the farmers Indian economy as it is a main source of income as large
will be benefited to make the planning appropriately with population of India depends on agriculture. Agriculture deals
the predicted results. The yield of the crop depends on the with the business of crops and also depends on some factors.
major factors like weather, soil, water, nitrogen, soil Some of the factors on which the agriculture is dependent
moisture, humidity, crop rotation, surface temperature upon are the type of soil, the climatic conditions, cultivation of
rain water etc. Predicting the crop yield deals with the crop, temperature, fertilizers used, harvesting, irrigation,
huge datasets along with lots of data analytics with are pesticides which are used and other factors. Sometimes there
applied on those datasets. Prediction of agriculture crop would also be a huge loss in agriculture due to the natural
yield is a job that requires the unification of knowledge calamities like drought, flood which would result in the loss of
from several areas such as data mining. Using data mining crop production and loss to farmers. The large numbers of
we extract knowledge of huge data using various farmers are committing suicide due to this reason and
algorithms/techniques. sometimes there would also be a pest attack which will again
result in the loss. Annually, farmers face a loss of 40% in total
A huge data coming from different parameters that is caused by either the damage of crops or the farmers
like weather, water, pH, temperature, soil quality etc are deprived of being paid the right price for their yield.
required for crop yield prediction .Prediction of crop Therefore, this results in low productivity. The crop yield in
yield is important for managing and storage of crops and agriculture is measured of the degree of yield per unit area that
data mining also focuses upon methodologies used for has been cultivated with the seeds that are produced from the
extracting the useful knowledge from data and there same cultivated crop. Predicting the crop yield immediately
various tools used for extracting knowledge from data. after the cultivation will be helping the farmers and the
But it is sad that farmers do not use any knowledge government bodies to decide for storage, the yield profit,
discovery process approach for their crop yield data. selling of crops, importing/exporting the crop. Government
Data mining is also helpful in agriculture for decision has taken few steps and is providing the messages and the
making. Our system aims at making farmers aware of information of the crop that is known as the agro vendor’s
the crop yield prediction by using various data mining information to farmers; also providing data related to soil
techniques and the result of the prediction as a message quality and the utility of the soil at each region. The yield that
to his phone. is derived from the crop will be impacting directly on the
National and International economic sectors of the country on
Keywords:- data mining, agriculture, prediction, mobile annual basis and also plays an important and significant role in
application. the food management. There is need for agricultural
institutions that educate the farmers with short-term courses
I. INTRODUCTION and workshops, update them with latest technologies and
hoping to obtain better and effective results. Forecasting the
India is the seventh largest country in the world and information of the crop before harvest is very useful for the
it’s an agriculture based country where most of its population farmers for future activities. There are many researchers who
derive their living and the source of income from the have developed statistical models for the yield of crop that
agriculture itself. India is practicing agriculture since many would suggest the pre-harvesting process and the crop yield
hundred years and it is the oldest countries to follow after harvesting. These industries use agriculture product as
agriculture even today. Due to globalization, the trends in the chemicals, seeds, poultry, fertilizer, livestock, food animal
agriculture are keep on changing .India is being ranked second feed, raw material, pesticide, paper. The accurate estimation of
in the farm output worldwide. Most of the economy in India is

IJISRT18AP602 www.ijisrt.com 431


Volume 3, Issue 4, April – 2018 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
crop production helps these companies to plan supply chain the redundant data will be removed. Finally that
decision in production. Till today by keeping only the redundant data will be normalized.
experience of the farmers was used, this included the random Feature Extraction: Here selection of attribute will
count of number of seeds, buds that plants have. Based on happen. It identifies and uses the relevant attribute from set of
these the farmer would predict the crop yield. Now days there data. The classifiers application processes the irrelevant and
are number of different applications emerging and the redundant data and then those data will be removed.
technologies for the farmers to make their lives easier. The
information technology has become more and more part of In [2] it follows some methodology
one’s daily life, which is helpful especially in the field of
agriculture. The Data mining is one of the processes of • Dataset Collection: The dataset is gathered from
discovering the previously unknown and the likely impressive Madurai district and in soil testing lab it has been tested for
patterns in the large datasets. Therefore the Data Mining is the some specific attribute of soil. In addition to that some general
process used for extracting information from the given set of
crop data is collected from online sources which are similar.
data and transforms it into a readable structure for further use
of data. Data Mining is a process used for discovering the The crops considered may include cotton, groundnut, millet,
patterns in the large data sets which are involving methods at pulses, sorghum, sugarcane, coriander, vegetables, banana,
which include the statistics , machine learning, artificial and paddy in this model. Texture, Depth, Drainage,
intelligence and database systems. Finally the ultimate goal of Permeability, soil colour, Ph, Water holding and Erosion are
data mining is prediction - and predictive data mining is the the parameters considered. These above mentioned soil
most common type of data mining and one that has the most parameters play an important role in extraction of soil
direct business applications. Data mining is the technique
nutrients and water of crop’s ability. For the satisfactory
which plays a vital role in the analysis of data. Data mining is
also computing process which is used for discovering patterns environment soil is must. The absorbed nutrients determine
in large data sets involving methods at the intersection of the water holding capacity. To hold onto the nutrients soil
artificial intelligence, machine learning, statistics, and texture will affect the soil’s ability. The availability of soil
database system. Unsupervised (clustering) and supervised nutrients is affected by the acidity or alkalinity level which is
(classifications) are two different types of learning methods an important variable. The penetration of roots is determined
which are implemented in the data mining. Clustering is the by water holding and drainage. For choosing a crop the
process of examining a collection of “data points,” and
following reasons are considered for the stated parameters.
grouping the data points into “clusters” according to some
distance measure.
• Ensemble technique for crop prediction: Ensemble is
II. RELEATED WORK one of the models of data mining to gather greater efficiency
and prediction this model combines the multiple models
The survey consists of study of various power than other model could alone achieve. This is also
methodologies that were adopted to predict the crop yield.
called as the Method of Committee or Combiners Model. In
In [1] using crop yield prediction system for this system they use the ensemble technique called Majority
enhancing the yield for better decisions and planning which voting technique. Base learners can be at least two.
can be planned out. They follow some step.
 Learners Used in the Model
• Selecting the agriculture field: First for the crop yield  Random Tree
prediction we need to select the agriculture field. It is same as decision tree. But the difference is for
• Selecting the crop: The crop which needs to be sown every split their will be random attribute subset is available.
should be selected. Nominal and numerical data both can be built for random
• Input: Consider the climatic parameters like rainfall, trees. Randomly chosen attributes can be considered for each
temperature, humidity and also some agronomical node. The size of the subset can be specified by subset ratio
variables like pesticides, nutrient contents N, P, K, and parameter.
soil etc. The parameters values will be taken as input.
• Pre-processing: It is nothing but noise removing process.  Chaid
The data mining to be a successful application it requires It stands for Chi-squared Automatic Interaction
Detection. It is based upon adjusted significance testing which
dataset which is huge. From the various resources the data is also a decision tree technique type. Chi-squared based
which is collected will be in raw form and that will cotain criterion is used in place of gain or information ratio. It has
some redundant, inconsistent, incomplete data. So here many advantages which uses multi way splits that include
highly interpretable and visual results.

IJISRT18AP602 www.ijisrt.com 432


Volume 3, Issue 4, April – 2018 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
 K-Nearest Neighbor C. System Design
For both classification and regression this can be
used. The cases available will be stored and new cases will be
classified according to similarity measure.

III. METHODOLOGY

A. Working Principle

• Gather the data related to the field.


• Compute the collected data with the database
• Communicate the data to the server along with
the necessary parameters like soil, climate etc.
• Get the result.
• Notification to the farmer about the yield prediction.

B. Work Flow

Login To The App Entering The Parameters

D. Exixting System
In existing systems there is no much technology used
for the detection or the prediction of the yield, based on the
previously collected data. This previously collected data will
include the several factors like soil, climate, rainfall and other
important features in that particular location. At present only
the type of the crop that can be grown in that particular area or
region is been told to the farmers and most the farmers fails to
go to the government bodies to take suggestions about the
Interfacing Db land, soil, and other climatic conditions that will be effecting
the growth of their crops in spite of all these facilities the
farmers fail to go because they find difficult to travel to far
place and no proper segments to support them therefore the
farmers will not have much knowledge about the crop yield in
advance.

E. Proposed System
The system that we are proposing here is the mobile
application where the farmers need to download the app on
Using Datamining Techniques
their smart phone and enter the required details about their
field and the data related to the crop they grow and for how
much area along with the climatic conditions ,rainfall and
other necessary factors and all these data will be compared
with the previously noted data and are compared using various
Final Yield Predicted data mining techniques then finally the compared results of
crop yield would be displayed on their smart phone.
Fig 1:- The system work flow

IJISRT18AP602 www.ijisrt.com 433


Volume 3, Issue 4, April – 2018 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165

REFERENCES

[1]. Yogesh Gandge, Sandhya, study in various data mining


techniques for crop yield prediction, 2017 International
Conference on Electrical, Electronics, Communication,
Computer and Optimization Techniques (ICEECCOT)
[2]. S.Pudumalar, E.Ramanujam,R.Harine Rajashree,
C.Kavya, T.Kiruthika, J.Nisha. Crop Recommendation
System for Precision Agriculture 2016 IEEE Eighth
International Conference on Advanced Computing
(ICoAC)978-1-5090-5888-4/16/$31.(2016) IEEE.
[3]. S.Nagini, Dr. T. V. Rajini Kanth, B.V.Kiranmayee,
Agriculture Yield Prediction Using Predictive Analytic
Techniques, 978-1-5090-5256-1/16/$31.00(2016)IEEE.
[4]. Monali Paul, Santosh K. Vishwakarma, Ashok Verma,
Analysis of Soil Behaviour and Prediction of Crop Yield
using Data Mining Approach,2015 International
Conference on Computational Intelligence and
Communication Networks.
[5]. Jharna Majumdar, Sneha Naraseeyappa and Shilpa
Ankalaki Analysis of agriculture data using data mining
techniques: application of big data Majumdar et al. J Big
Data (2017).
[6]. A predictive modeling approach for improving paddy
crop productivity using Data Mining Approach, Turk J
Elec Eng & Comp Science (2017).
[7]. Effects of temperature and rainfall on paddy yield using
data mining .kuljit kaur,kanwalpreet singh.987-1-5090-
3519-9/17/$31.00 (2017)IEEE.
[8]. Niteta Gandhi, Liesa Armstrong, Applying data mining
technology to predict yield of rice in humid subtropical
climatic zone of india,978-9-3805-4421-
2/16/$31.00(2016)IEEE

IJISRT18AP602 www.ijisrt.com 434

You might also like