USING TEXT MINING TO CLASSIFY RESEARCH PAPERS.

Recently, the volume of scientific literature has grown rapidly raising an imminent question about its storage and organization. Many research papers are often available only through the websites of the relevant scientific journals. It is an essential problem when different classification codes are used in order to organize these papers or when specific categorization in a certain scientific field is missing. This leads to unnecessary complications in the researchers' aims who want to quickly and easily find literature on a specific topic among the large amount of scientific publications. Simultaneously, the research interest related to the mechanisms of natural language processing is growing because much of the information they work with is unstructured and in the form of plain text. In order to improve and automate the process of organizing and classifying scientific papers we propose an approach based on the technology for natural language processing. This applies the methods of supervised machine learning and two specific algorithms for text categorization - Support Vector Machines (SVM) and Naive Bayes (NB). The proposed approach classifies the scientific literature according to its contents. To successfully execute our scientific research, we used over 200 papers, published in the last four years in the journal "Izvestiya", which is issued by the University of Economics - Varna. The articles explore different topic areas and are written in English. The experiments were conducted with the software product RapidMiner. [ABSTRACT FROM AUTHOR]