A comparative study of data mining techniques for agriculture crop price prediction

Abstract

Agriculture crop prices forecasting is a very interesting and high challenging process as it is fully dependent on upcoming production in entire country. Recently most available application is designed for price analysis rather than price forecasting. In India agriculture production, when it is calculated per farmer, it is very instable is there compare to rest of the world, when compared to individual farmer in various countries with stable environment, and without providing sufficient MSP it will not benefit farmers and agriculture fraternity. If the farmers and agriculture fraternity get an access to appropriate crop prices, then poverty can be reduced in India. In advanced agriculture development, a large quantity of data is generated from the agriculture commodity market. Agriculture has a large amount of data, however regrettably, most of this data is not extracted to find out unseen information in data— crop price forecast is more beneficial to the farmers and agriculture fraternity to take proper and timely decisions. According to the output of process, Advanced data mining techniques play a pivotal role in analysis to discover a hidden pattern in data. Performance of data mining techniques is compared with past crop prices, weather, current market prices, stock availability and the upcoming production of the crop in recent years. The data mining that is a regression analysis, Tracking Patterns, Cluster Analysis, and visualization techniques are used to create an inventive representation to predict the agricultural crop prices.