• M.INDIRA
• Dr. S.JAYASANKARI

Data mining is the process of analyzing the data from the large database for extracting the valuable information with the relevant features. Feature selection is a preprocessing step before the classification to improve the mining performance by eradicating the irrelevant features and choosing the relevant features from the database. The several feature selection is done using different data mining techniques but it has high computational complexity. In order to improve the feature selection and classification, a Pearson Chi-Square Based Kernel Least Square Support Vector Classifier (PCS-KLSSVC) is introduced. The PCS-KLSSVC includes two processes namely feature selection and classification for identifying the seed growth in the agriculture field. At first, the feature selection is performed using Pearson chi-squared hypothesis test. Based on the score value, relevant features are selected by removing the irrelevant features for performing classification. This helps to improve feature selection rate and minimizes the classification time. Secondly, the Kernelized Least Square Support Vector Classifier is used for performing the classification process to predict the seed growth. The classifier constructs an optimal and marginal hyperplane for improving the classification accuracy and minimizes the error rate. Experimental evaluation of proposed PCS-KLSSVC and existing methods are carried out with different factors such as feature selection rate, classification accuracy, classification time and space complexity with respect to a number of features and data points. The experimental results reported that the proposed PCS-KLSSVC obtains high classification accuracy and feature selection rate with minimum time as well as space complexity.

Keywords: Data mining, Pearson chi-squared hypothesis test, feature selection, Kernelized Least Square Support Vector Classifier, hyperplane, seed growth.

"ENSEMBLED PEARSON CHI-SQUARE FEATURE SELECTION AND KERNEL LEAST SQUARE SUPPORT VECTOR CLASSIFIER FOR AGRICULTURE SEED GROWTH", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.5, Issue 9, page no.391-400, September-2018, Available :http://www.jetir.org/papers/JETIR1809207.pdf

"ENSEMBLED PEARSON CHI-SQUARE FEATURE SELECTION AND KERNEL LEAST SQUARE SUPPORT VECTOR CLASSIFIER FOR AGRICULTURE SEED GROWTH", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.5, Issue 9, page no. pp391-400, September-2018, Available at : http://www.jetir.org/papers/JETIR1809207.pdf