Downloads: 11 | Views: 245 | Weekly Hits: ⮙1 | Monthly Hits: ⮙1
Research Paper | Computer Science | India | Volume 11 Issue 11, November 2022 | Popularity: 5.3 / 10
Analysis of Long Short Term Memory (LSTM) Network for Rice Crop Yield Prediction
Manasvin A
Abstract: Agriculture is a crucial aspect of India, both in terms of economy and culture. It employs more than 50% of the Indian workforce. There is a need to provide a helpful guidance to the farmers with regards to what their yield is going to be. In the modern economic situation, farmers are under a lot of stress to better plan their schedules and harvests. Having a tool that can provide a estimation or prediction on the crop yield can help the farmers make predictions and also observe the trends. Then there is also the aspect that the climate is experiencing striking and impactful changes due to rising pollution, use of harmful energy sources, etc. Traditional yield predictors tend to only consider discrete current climate parameters, and not using climate data gathered up to that point in time. Past studies have shown that the adverse negative effects of climate change are going to be experienced by developing countries. Hence, there is a need to take into account the past climate trends when predicting crop yields. This paper explores the applicability of Long Short Term Memory (LSTM) networks, a form of Recurrent Neural Network (RNN), using time series crop yield data in prediction of rice yields in the states of Uttar Pradesh, Bihar and Karnataka. Possible improvements and extensions that can be applied to the model to improve its accuracy are also explored.
Keywords: Rice yield, time series, Long Short-Term Memory
Edition: Volume 11 Issue 11, November 2022
Pages: 1338 - 1342
DOI: https://www.doi.org/10.21275/SR221125154150
Please Disable the Pop-Up Blocker of Web Browser
Verification Code will appear in 2 Seconds ... Wait