Deep Learning with Particle Swarm Based Hyper Parameter Tuning Based Crop Recommendation for Better Crop Yield for Precision Agriculture

R Rangaraj,K Mythili

doi:10.17485/ijst/v14i17.450

Abstract

The common difficulty present among the Indian farmers is that they don’t opt for the proper crop based on their soil necessities. Because of this productivity is affected. This problem of the farmers has been solved through precision agriculture. This method is characterized by a crop database collected from the farm, crop provided by agricultural experts and given to recommendation system it will use the collect data and do deep learning model as learners to recommend a crop for site specific parameter with high accuracy and efficiency. Objectives: The comprehensive objective of the model is to deliver direct advisory services to even the smallest farmer at the level of his/her smallest plot of crop, using the most accessible technologies using deep learning. It is a recommender model built using a classifier and an optimization of the classifier. Based on appropriate parameters, the system recommends crops as technology based crop recommendation system in agricultural decisions can be of great help to farmers in increasing their crop yields or cultivating suitable crops based on their land characteristics and climatic parameters. Methods: Existing MLTs (Machine Learning Techniques) have been found to be expensive and fail to scale in locally collected samples, so recent techniques of deep learning method is adopted in this work. With this motivation, historical data on crop productions and climate data is used in this work which is then pre-processed followed by predictions using a Modified DNN (Deep Neural Networks) and PSO (Particle Swarm Optimization) called PSO-MDNN for recommending crop cultivation. This work uses L2 regularizations for reducing weights in matrices assuming. Findings: This work proposed MDNN where the weight matrices are calculated with L2 regularization and PSO utilized to tune the hyper parameters of MDNN and its network structure to improve the prediction accuracy. Novelty: MDNN can produce to simpler models while working with weights that can be optimized through PSO. PSO-MDNN submodels produced from datasets and this proposed models gets adapted to data patterns if featuring datasets. This proposed work also exhibits promising outcomes in predicting crop yields using DLTs. Keywords: Crop recommendation system; Crop yield; Precision agriculture; Deep learning; Modified Deep Neural network and Particle Swam Optimization

Highlights

In recent times, forecasting of crop productivity at the within-field level has increased
As there exists a nonlinear relationship between crop yield and the factors influencing crop, machine learning techniques might be efficient for yield predictions
The proposed technique is benchmarked with existing research techniques(10) (11) like Decision Tree, KNN, R-Forest, Neu-net and proposed Particle Swarm Optimization algorithm (PSO) MDNN using the metrics of accuracy, precision, recall and f-measure for evaluations

Summary

Introduction

In recent times, forecasting of crop productivity at the within-field level has increased. The most influencing factor for crop productivity is weather conditions. If the weather based prediction is made more precise, farmers can be alerted well in advance so that the major loss can be mitigated and would be helpful for economic growth. Predicting crop yield can facilitate the farmers to have a better vision on cultivation of seasonal crop and its scheduling(1). Crop selector could be applicable for minimize losses when unfavorable conditions may occur and this selector could be used to maximize crop yield rate when potential exists for favorable growing conditions(2). It is necessary to simulate & predict the crop yield before cultivation for efficient crop management and expected outcome. As there exists a nonlinear relationship between crop yield and the factors influencing crop, machine learning techniques might be efficient for yield predictions

Objectives

Methods

Results

Conclusion