Crop Yield Prediction Using Random Forest Algorithm

Aim:

      Predict the crop type and price of the crop using machine learning methodology with accurate results.

Abstract

       India is the land of agriculture and it is the major source of economy.70% of Indian population directly relies on agriculture. The common problem existing among the young Indian farmers is to choose the right crop based on the location, humidity, temperature, rainfall. Due to this, they face a serious setback in productivity. Our work proposes to help farmers determine the predict crop type and price by doing analysis on its various parameters and to suggest crops based on the results obtained. The system uses the Classification algorithm of Random Forest to improve the efficiency of Crop Recommendation System. The system maps the location, temperature, humidity, pH value and rainfall to predict the list of suitable crops for the soil and it also provides cost of the crop. Hence it leaves upon the user to decide on the crop to be sown. Thus, the system helps to provide knowledge to the dilettante farmers.

Existing System:

       Now a day’s, dilettante farmers are hard to understand the cultivation process, crop type, climate change, etc. Farming is that the spine for every nation’s economy. Future agriculture depends on dilettante formers. But, the new farmers have low level knowledge in this field. So, Machine learning help to solve this type of problems. In Existing system they provide soil type and crop using Random Forest algorithm. But everyone can able to find a soil type easily. So, we need to predict the crop type and predict the crop price based on Machine learning technology.

Crop Problem Statement:

• Crop yield prediction is an important agricultural problem. The Agricultural yield primarily depends on weather conditions (rain, temperature, etc), pesticides.
• Accurate information about history of crop yield is important for making decisions related to agricultural risk management and future predictions.
• Now, some new formers or students are willing to work in agriculture means they don’t known about that field.
• So, we collected the past data and using the machine learning algorithm to predict the crop and soil parameters based on our previous model.

Proposed System:

            The system prepared predict major crops yield in a particular district in Tamil Nadu. The client on their first login has to register themselves on the application on android phone .Once the user logins into the system he gets all the access for predicting crop yield and using the input such as location, temperature, pH value, rainfall and humidity depends on their forming land environment. After submitting the inputs, it’s redirect into Firebase. The firebase is an intermediate between user input and trained data set. The input goes to the trained data, where it processes random forest algorithm to predict crop and price. After the prediction, the predicted value passes to the fire base. That firebase gives the predict value to the user on android application.

Advantage

• The proposed system provides connectivity to new farmers via a Web 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.
• Crop prediction reduces risk of financial loss due unfavorable conditions. And to reduce risk of market fluctuations, mechanism of farming, growing expensive crops.