Implementation of K-Means Clustering Technique in Banana Production of Tamil Nadu, India

Abstract

Aim: The main objectives of this study are to make use of the K-Means clustering approach to cluster the Banana data and to assist with crop yield prediction.
 Study Design: One of the methods of Big Data Analytics K-Means clustering is usedto cluster the data set.
 Place and Duration of Study: So far, the period 2010-2020, time series data were collected from the season and crop report, Directorate of Economics and Statistics, Chennai.
 Methodology: The horticulture industry has a significant impact on India's economic development. In the globe, after China, India ranks second in terms of fruit and vegetable production. Compare to the various fruits Mango and banana are one of the most abundant fruits in India. So, the Banana dataset were collected and dataset were clustered using the K-Means clustering technique and the optimum number of clusters were identify using the elbow approach.
 Results: According to these results from this study, there is positive relationship between the Area, Soil moisture, Maximum Temperature, Relative Humidity and negative relationship between Rainfall, Wind Speed and Minimum temperature related Banana production. Using K-Means clustering it divides the given dataset into three clusters in which cluster 3 contains high Banana production afterwards two and one.
 Conclusion: The selection of the most productive clusters is going to tell farmers on where to focus their efforts while planting crops in order to enhance productivity and crop production.