Inspiration

Farming is one of the major sectors that influences a country’s economic growth.

In, a country like India, the majority of the population is dependent on agriculture for their livelihood. Many new technologies, such as Machine Learning and Deep Learning, are being implemented into agriculture so that it is easier for farmers to grow and maximize their yield.

Considering the present system including manual counting, climate-smart pest management, and satellite imagery, the result obtained isn’t really accurate. So in this project, we have mainly focused on recommending the crop and predicting the yield of the crop by applying various machine learning techniques.

What it does

Crop Recommendation - In this, we had built a model in which users can provide the soil data from their side and the model will predict which crop should the user grow. We have applied various machine learning algorithms including Random Forest, Naive Bayes, XGBoost, Logistic Regression, and Decision Tree. Among all of these XGBoost gave us an accuracy of 99.35% following RF and NB of 99.09%.
Crop Yield Prediction - In this, we had built a model that will predict the statewise crop production in India. The dataset had the features like state, crop year, season, area, and crop. Using this we predicted how much production we will get in a particular year.

How we built it

Using various Machine learning algorithms like Random Forest, Logistic Regression, XGBoost, SVM. we build a model that will predict the crop one should grow on their farm. Also, we have predicted the production of crops in India based on some parameters. Using the "IBM Z" platform we have explored the dataset which we got from "data.world".

Challenges we ran into

The dataset was not so clean. It was containing too many null values and many categorical variables. So we had to clean the dataset first. After that, we had to study the impact of each variable on crop production.

Accomplishments that we're proud of

We are aiming to help the farmers to increase their income with the help of data-driven decisions.

What we learned

Along with the data science techniques we have used, we have also learned how the agriculture sector works and how the farmers decide which crop should they grow in a particular season.

What's next for FarmerFriend

We are planning to convert it into a web application so that the users can directly assess it. Also, we planning to whether APIs so that the result would be more accurate. Also, we are planning to recommend water management so we can use water efficiently.