10 IV April 2022

https://doi.org/10.22214/ijraset.2022.41381
 International Journal for Research in Applied Science & Engineering Technology (IJRASET)
ISSN: 2321-9653; IC Value: 45.98; SJ Impact Factor: 7.538
Volume 10 Issue IV Apr 2022- Available at www.ijraset.com

Agriculture Crop Yield Prediction Using Machine

Learning
Firdous Hina1, Dr. Mohd. Tahseenul Hasan2
1, 2
Anjuman College of Engineering and Technology, Sadar, Nagpur-17

Abstract: In our suggested system, we employed a vast dataset that included all of India's states, whereas in the old system, just a
single state was considered. These suggestions may be extracted and used to educate the farmers. The farmer can have a better
understanding of the crops to cultivate by using a pictorial depiction. Machine Learning Techniques develops a well-defined
model with the data and helps us to attain predictions. Agricultural issues like crop prediction, rotation, water requirement,
fertilizer requirement and protection can be solved. Due to the variable climatic factors of the environment, there is a necessity to
have a efficient technique to facilitate the crop cultivation and to lend a hand to the farmers in their production and
management. This may help upcoming agriculturalists to have a better agriculture. A system of recommendations can be
provided to a farmer to help them in crop cultivation with the help of data mining. To implement such an approach, crops are
recommended based on its climatic factors and quantity. Data Analytics paves a way to evolve useful extraction from agricultural
database. Crop Dataset has been analyzed and recommendation of crops is done based on productivity and season
Keywords: Machine learning, Agriculture techniques, crop predictions

I. INTRODUCTION
Agricultural unpredictability, due to changing temperature season soil parameters are the factors which reduces productivity. More
production should be gained with a larger population and area, however this is impossible to achieve. Farmers used to rely
on word-of-mouth, but owing to climate reasons, they can no longer do so. Agricultural elements and characteristics are used to
provide data that may be used to learn more about Agri-facts. AgricultureSciences presents key highlights in the IT industry to assist
farmers with good agricultural information. In the current situation, the ability to apply new technical approaches to the sector of
agriculture is desirable. MachineLearning Techniques use data to create a well-defined model that may be used to make predictions.
Crop forecast, rotation, water requirements, fertilizer requirements, and crop protection are all challenges that may be resolved. All
learning takes place in the brain of machine learning. The machine learns in the same way that humans do. Humans gain knowledge
from their experiences. The more information we have, the easier it is to make predictions. By analogy, the chances of success are
lower in an unknown circumstance than in a known situation. Machines are taught in the same way. The machine examines an
example in order to produce an accurate forecast. When we give the machine a comparable scenario, it can predict what would
happen. However, much like a person, if the computer is fed an example that has never been seen before, it has trouble predicting
what will happen next.

Learning and inference are the two main goals of machine learning. First and foremost, the machine learns by recognising patterns.
Machine Learning programmes have a clear life cycle that may be stated as follows:
1) Formulate a query
2) Gather information
3) Visualize data
4) Train the algorithm
5) Put the algorithm to the test
6) Gather feedback
7) Fine-tune the algorithm
8) Repeat steps 4–7 until the results are satisfactory.
9) Make a forecast using the model.
Once the algorithm has mastered generating the correct conclusions, it applies what it has learned to fresh data sets.. To collect and
process data, technologies such as blockchain, IoT, machine learning, deep learning, cloud computing, and edge computing can be
used. Computer vision, machine learning, and IoT applications will assist farmers and related domains enhance production, improve
quality, and ultimately increase profitability.

©IJRASET: All Rights are Reserved | SJ Impact Factor 7.538 | ISRA Journal Impact Factor 7.894 | 910
 International Journal for Research in Applied Science & Engineering Technology (IJRASET)
ISSN: 2321-9653; IC Value: 45.98; SJ Impact Factor: 7.538
Volume 10 Issue IV Apr 2022- Available at www.ijraset.com

Pre-harvesting, harvesting, and post-harvesting are the three major categories of agricultural activity. Machine learning
advancements have aided in boosting agricultural gains. Machine learning is a recent technology that is assisting farmers in
reducing farming losses by offering detailed crop recommendations and insights.. Machine learning has by far been the most helpful
and an innovative method that is recommended to be used in agriculture based applications. A farmer can receive an accurate
estimate of harvestable versus non-harvestable acres on a particular day by logging into a customised dashboard on a computer or
tablet using machine learning technology. Harvestable crop weight and maturity can also be assessed and projected. Additionally,
crops can be examined both before and after harvest for the presence of attractive traits, level of damage (if applicable), nutritional
makeup, and other factors that may effect the ultimate viable yield and product pricing using a range of technologies, including
image analysis. Machine learning (ML) approaches are used in a number of businesses, from anticipating consumer phone usage to
analysing customer behaviour in supermarkets (Ayodele, 2010). (Witten et al., 2016). Machine learning has been used in agriculture
for some years (McQueen et al., 1995). Crop production prediction is one of the most challenging challenges in precision
agriculture, and various models have been proposed and verified thus far. This task involves the use of numerous datasets since crop
production is impacted by a number of factors such as climate, weather, soil, fertiliser use, and seed type (Xu et al., 2019). This
implies that estimating agricultural output is not a simple process, but rather a succession of complex steps. Crop yield prediction
programmes can already estimate actual yields with reasonable accuracy, although better yield prediction performance is still
sought. Researchers have helped to solve the problem of climate change's influence on agricultural output in a variety of ways. In
this section, we provide a quick assessment of the work of colleagues and attempt to critically analyse it in order to define the area
of future study. We will continue our investigation as part of our research .Its applications can be divided into four major categories:
a) Crop management
b) Livestock management
c) Water management
d) Soil management
II. RELATED WORKS
LIsmail et al. (2018) [1] created a system for predicting a country's crisis readiness. Climate change is being studied using machine
learning. Southeast Asia is the focus of the study. The processes involved in calculating the predictive index include data collecting,
data training, data testing, and index computation. An index can be used for prediction, index validation, and index visualisation,
among other things. The study is being carried out as a prophylactic measure. The regions will be notified, and the susceptible index
will be checked using deep learning.
III. PROPOSED WORK
Crop yield forecast is a critical challenge for decision-makers at all levels, including national and regional (e.g., EU) decision-
making. Farmers can use an accurate crop production prediction model to assist them determine what to grow and when to grow it
Crop yield forecasting is a difficult task for decision-makers at all levels, including at the national and regional (e.g., EU) levels.
Farmers may use a crop production prediction model to help them decide what to plant and when to plant it. Crop production
forecasting may be done in a number of different methods. This review paper looks at what has been written on the use of machine
learning in agricultural yield prediction in the literature. The fact that the publication is a survey or typical review piece is one of the
exclusion criteria in our analysis of the retrieved publications. Those articles that were left out are really part of a larger project
discussed in this section.
IV. MACHINE LEARNING IMPLEMENTATION

Fig i Machine Learning block diagram

©IJRASET: All Rights are Reserved | SJ Impact Factor 7.538 | ISRA Journal Impact Factor 7.894 | 911
 International Journal for Research in Applied Science & Engineering Technology (IJRASET)
ISSN: 2321-9653; IC Value: 45.98; SJ Impact Factor: 7.538
Volume 10 Issue IV Apr 2022- Available at www.ijraset.com

Python Libraries required for implemention

1) Numpy
2) Pandas
3) Sci-kit Learn
4) Matplotlib

A. Pre-processing of Data
The technique we choose to collect data is determined on the type of project we wish to build. For example, if we want to construct
an ML project that leverages real-time data, we may use data from numerous sensors to create an IoT system. The data set can
originate from a variety of sources, including a file, a database, a sensor, and many others, but it cannot be used for analysis right
away because it may contain a large amount of missing data, exceptionally large values, disorganised text data, or noisy data. As a
result, Data Preparation has been performed in order to address this problem. Data pre-processing is one of the most important steps
in machine learning. It's the most important stage in making machine learning models more accurate. In machine learning, there is
an 80/20 rule. Every data scientist should spend 80% of their time pre-processing data and 20% of their time on actual analysis.

V. DATA CLASSIFICATION
When the goal variable is categorical (i.e., the output may be classified into classes – it belongs to Class A, B, or something else), a
classification problem arises.

Fig ii: Classification techniques

VI. ALGORITHMS FOR CLASSIFICATION

A. K-Nearest Neighbor
B. Naive Bayes
C. Decision Trees/Random Forest
D. Support Vector Machine
E. Logistic Regression

We split a model into three sections to start training it: 'Training data,' 'Validation data,' and 'Testing data.' You train the classifier
with a 'training data set,' tweak the parameters with a 'validation set,' and evaluate the classifier's performance with a 'unseen test
data set.' It's worth remembering that during training, the classifier only has access to the training and/or validation sets. When the
classifier is being trained, the test data set must not be used. Only while the classifier is being evaluated will the test set be
available..

©IJRASET: All Rights are Reserved | SJ Impact Factor 7.538 | ISRA Journal Impact Factor 7.894 | 912
 International Journal for Research in Applied Science & Engineering Technology (IJRASET)
ISSN: 2321-9653; IC Value: 45.98; SJ Impact Factor: 7.538
Volume 10 Issue IV Apr 2022- Available at www.ijraset.com

Fig iii: Algorithms for classification

The project's problem statement is to use a Decision Tree Classifier to propose crops to farmers. The main method of this project is
to preprocess the data supplied to us, then utilise it to construct the model for the backend and link it to the UI interface using flask
to display the entire and final product. In our suggested system, we employed a vast dataset that included all of India's states,
whereas in the old system, just a single state was considered. These suggestions may be extracted and used to educate the farmers.
The farmer can have a better understanding of the crops to cultivate by using a pictorial depiction. There are 821 unique data points
in the dataset. The dataset has 14 columns, which are explained below.

1) States: India's total number of states

2) Ground Water: Total ground water level
3) Rainfall: rainfall in millimetres
4) Temperature: the temperature is measured in degrees Celsius.
5) Soil type: There are a variety of soil kinds.
6) Season: When is the best time to plant crops?
7) Crops: Different types of crops
8) Fertilizers needed: Types of Fertilizers Needed
9) Cultivation cost: Total Cultivation Cost
10) Revenues forecasted: Total revenue forecasted
11) Seeds per hectare amount: seeds per hectare quantity
12) Culture duration: amount of days for cultivation length
13) Crop demand (nineteenth): crop demand (ninth): crop demand (ninth
14) Crops for mixed cropping: which crops can be cropped together

©IJRASET: All Rights are Reserved | SJ Impact Factor 7.538 | ISRA Journal Impact Factor 7.894 | 913
 International Journal for Research in Applied Science & Engineering Technology (IJRASET)
ISSN: 2321-9653; IC Value: 45.98; SJ Impact Factor: 7.538
Volume 10 Issue IV Apr 2022- Available at www.ijraset.com

Fig iv: Algorithms for classification

Gather information and prepare it for training. Clean anything that may require it (remove duplicates, correct errors, deal with
missing values, normalization, data type conversions, etc.) Randomize data to remove the impacts of the sequence in which we
gathered and/or otherwise prepared our data. Perform further exploratory analysis, such as visualising data to assist uncover
meaningful associations between variables or class imbalances (bias alert. Sets for training and evaluation have been created.

Fig v: Flowchart

VII. CONCLUSION
The importance of crop management was extensively addressed in this research. Farmers require modern technologies to help them
raise their crops. Agriculturists can be notified about accurate crop predictions on a timely basis. The agricultural factors were
analyzed using a variety of Machine Learning approaches. The features chosen are determined by the dataset's availability and the
research's goal. According to studies, models with more characteristics may not always deliver the highest yield prediction
performance. Models with more and fewer features should be evaluated to discover the best performing model. Several algorithms
have been employed in various research. The findings reveal that while no definitive conclusion can be taken about which model is
the best, they do show that some machine learning models are utilized more frequently than others. With the incorporated model we
found an average accuracy of 95%.

©IJRASET: All Rights are Reserved | SJ Impact Factor 7.538 | ISRA Journal Impact Factor 7.894 | 914
 International Journal for Research in Applied Science & Engineering Technology (IJRASET)
ISSN: 2321-9653; IC Value: 45.98; SJ Impact Factor: 7.538
Volume 10 Issue IV Apr 2022- Available at www.ijraset.com

REFERENCES
[1] Shreya S. Bhanose, Kalyani A. Bogawar (2016) “Crop And YieldPrediction Model”, International Journal of Advance ScientificResearch and Engineering
Trends, Volume 1,Issue 1, April 2016
[2] A.T.M.S. Ahamed, N.T. Mahmood, N. Hossain, M.T. Kabir, K. Das, F. Rahman, R.M. Rahman Applying data mining techniques to predict annual yield of
major crops and recommend planting different crops in different districts in Bangladesh 2015 IEEE/ACIS
[3] I. Ahmad, U. Saeed, M. Fahad, A. Ullah, , A. Ahmad, J. Judge Yield forecasting of spring maize using remote sensing and crop modeling (10) (2018)
[4] I. Ali, F. Cawkwell, E. Dwyer, S. GreenModeling managed grassland biomass estimation by using multitemporal remote sensing data—a machine
learning approachIEEE J. 10 (7) (2017),
[5] Cheng, H.; Damerow, L.; Sun, Y.; Blanke, M. Early Yield Prediction Using Image Analysis of Apple Fruit and Tree Canopy Features with Neural Networks. J.
Imaging 2017,
[6] Chen, Y.; Lee, W.S.; Gan, H.; Peres, N.; Fraisse, C.; Zhang, Y.; He, Y. Strawberry Yield Prediction Based on a Deep Neural Network Using High-Resolution
Aerial Orthoimages. Remote Sens. 2019,
[7] Črtomir, R., Urška, C., Stanislav, T. et al. Application of Neural Networks and Image Visualization for Early Forecast of Apple Yield. Erwerbs-Obstbau 54,
69–76 (2012).
[8] Duo Attention with Deep Learning on Tomato Yield Prediction and Factor InterpretationPRICAI 2019: Trends in Artificial Intelligence, 2019,
[9] D. Elavarasan and P. M. D. Vincent, "Crop Yield Prediction Using Deep Reinforcement Learning Model for Sustainable Agrarian Applications," in IEEE
Access, vol. 8, pp. 86886-86901, 2020,
[10] Tripathy, A. K., et al.(2011) "Data mining and wireless sensornetwork for agriculture pest/disease predictions." Information andCommunication Technologies
(WICT), 2011 World Congress on.IEEE.
[11] PriyankaP.Chandak (2017),” Smart Farming System Using DataMining”, International Journal of Applied Engineering Research,Volume 12,

©IJRASET: All Rights are Reserved | SJ Impact Factor 7.538 | ISRA Journal Impact Factor 7.894 | 915