Sat - 80.Pdf - Crop Yield Prediction Based On Indian Agriculture Using Stacking Algorithm
Sat - 80.Pdf - Crop Yield Prediction Based On Indian Agriculture Using Stacking Algorithm
I would like to express my sincere and deep sense of gratitude to my Project Guide
Mr.S.DHAMODARAN.,Ph.D., for his valuable guidance, suggestions and constant
encouragement paved way for the successful completion of my project work.
I wish to express my thanks to all Teaching and Non-teaching staff members of the Department
of COMPUTER SCIENCE AND ENGINEERING who were helpful in many ways for the
completion of the project.
5
ABSTRACT
6
TABLE OF CONTENTS
ABSTRACT 6
LIST OF FIGURES 9
10
LIST OF TABLES
LIST OF ABBREVIATIONS 11
1 INTRODUCTION
1.1 GENERAL 2
1.2 OBJECTIVE 3
3 METHODOLOGIES 10
7
6
RESULTS AND DISCUSSION, PERFORMANCE 47
7 SUMMARY AND CONCLUSIONS 68
8 APPENDICES 54-67
9
REFERENCES 69
8
LIST OF FIGURES
9
ABSTRACT
Agriculture and its allied sectors are undoubtedly the largest providers of
livelihoods in rural India. The agriculture sector is also a significant contributor
factor to the country’s Gross Domestic Product (GDP). Blessing to the country is
the overwhelming size of the agricultural sector. However, regrettable is the yield
per hectare of crops in comparison to international standards. This is one of the
possible causes for a higher suicide rate among marginal farmers in India. This
paper proposes a viable and user-friendly yield prediction system for the farmers.
The proposed system provides connectivity to farmers via a mobile application.
The user provides the area & soil type as input. Machine learning algorithms
allow choosing the most profitable crop list or predicting the crop yield for a user-
selected crop. To predict the crop yield, selected Machine Learning algorithms
such as Support Vector Machine (SVM), and K-Nearest Neighbour (KNN) are
used. Among them, the Random Forest showed the best results with 95%
accuracy. Additionally, the system also suggests the best time to use the fertilizers
to boost up the yield.
1
CHAPTER 1
INTRODUCTION
1.1 GENERAL
2
1.2 OBJECTIVE
The proposed model provides crop selection based on economic and
environmental conditions, and benefit to maximize the crop yield that will
subsequently help to meet the increasing demand for the country's food supplies.
The proposed model predicts the crop yield by studying factors such as
rainfall,temperature, area, season, soil type etc.
The system also helps to determine the best time to use fertilizers. The existing
system which recommends crop yield is either hardware-based being costly to
maintain, or not easily accessible.
temperature, area, season, soil type etc. The system also helps to determine the
best time to use fertilizers. The existing system which recommends crop yield is
either hardware-based being costly to maintain, or not easily accessible.
3
CHAPTER 2
LITERATURE SURVEY
To keep up nutrition levels in the soil in case of deficiency, fertilizers are added
to soil. The standard issue existing among the Indian agriculturists choose
approximate amount of fertilizers and add them manually. Excess or deficient
extension of fertilizers can harm the plants life and reduce the yield. This paper
gives overview of various data mining frameworks used on cultivating soil dataset
for fertilizer recommendation.
This paper communicates the idea regarding the making of AgroNutri an android
application that helps in conveying the harvest particular fertilizer amount to be
applied. The idea is to calculate the measure of NPK composts to be applied
depend on the blanked proposal of the crop of interest. This application works
4
depend on the product chosen by the farmer and that is taken as input, thus
providing the farmers. The future scope of the AgroNutri is that GPRS can be
included so that according to location nutrients are suggested. Further this
application would be incorporated as a piece of the accuracy agriculture wherein
sensors can be utilized to discover the measure of NPK present in the dirt and that
sum can be deducted from the suggestion and giving us the exact measure of
supplements to be added.
Agriculture is a field that has been lacking from adaption of technologies and their
advancements. Indian agriculturists should be up to the mark with the universal
procedures. Machine learning is a native concept that can be applied to every field
on all inputs and outputs. It has effectively settled its ability over ordinary
calculations of software engineering and measurements. Machine learning
calculations have improved the exactness of artificial intelligence machines
including sensor based frameworks utilized in accuracy farming. This paper has
evaluated the different uses of machine learning in the farming area. It
additionally gives a knowledge into the inconveniences looked by Indian farmers
and how they can be resolved using these procedures.
Throughout the following decades humanity will request more food from less land
and water assets. This investigation evaluates the food production effects of four
elective advancement situations from the Millennium Ecosystem Assessment and
the Special Report on Emission Scenarios. partialy and jointly considered are land
and water supply impacts from population development, and specialized change,
and forests and agriculture demand request shifts from population development
5
controller for the threephase induction machine adjustable speed system has been
designed. The proposed controller is not only of the advantages that FIS does not
depend on the plant model, strong robustness, and adaptive self-learning ability,
but also learning ability and generalization performance of SVM. The designed
processes of SVM-FIS, MDFP, and SVMI algorithms have been described in
details. Simulation results show the feasibility, correctness and effectiveness of
the proposed control strategy, such as the excellent static and dynamic
performances, and strong anti-interference ability.