Estimating the text preprocessing algorithms for sentence classification

Abstract

Estimating the text preprocessing algorithms in machine learning classification task helps to select the most accurate combination of word embedding text processing algorithms to map the test data samples to appropriate labels. Increasing the accuracy in classification of text data improves afterwards the response to queries in search engines and gives more relevant results. This work includes several preprocess text data techniques and their combinations for following sentence classification by using support vector machine (SVM). I.