Agriculture Data Feature Extraction Using Sequential Pattern Mining and Random Forest (SPM-RF) Data Mining Method

Abstract

Feature extraction assumes a critical part in analyzing agricultural datasets and acquiring experiences for decision-making. In this paper we propose a methodology that consolidates Sequential Pattern Mining (SPM) and Random Forest (RF) techniques for feature extraction in agricultural datasets. The SPM RF approach uses the temporal request of occasions in agricultural information to extract regular sequential patterns, which are then used to determine enlightening features. These features are thusly incorporated with the first dataset to make a feature-improved dataset. The Random Forest calculation is utilized to prepare a prescient model utilizing the improved features. Experimental results show the viability of the proposed SPM RF approach in extracting significant features from agricultural datasets, empowering further developed prediction in the agricultural area.