Classification of text documents using genetic algorithm and K-nearest neighbors

Text Mining is a branch of data mining that aims to obtain previously unknown information from large quantities of textual data. Classification, clustering and estimation are some important piece of Text Mining. An important part of a successful Text Mining is the successful classification process. Dimension reduction is usually performed to improve the success and efficiency of the classification system. In this study, the dimension reduction process was performed in the classification of text documents. Two methods have been used for this. One of them is feature selection and the other is feature extraction. Principial Component Analysis method is used for feature extraction. Weighting with coefficients is used for selected features after feature selection. Genetic Algorithm is used for the feature selection phase and for the selection of the best coefficients after feature extraction. According to the experimental results, the feature selection partially reduced the classification success. Feature extraction and coefficient weighting added after this step significantly increased the classification success.