Sentiment Analysis for Diagnostic Purposes

Abstract

The natural language processing has been the subject of numerous studies over the last decade. They have focused on the various stages of text processing, from the prepared text, through vectorization, to the final understanding. One of the greatest challenges nowadays is to understand the emotions that are express in written text. This is all the more difficult because sometimes people are also unable to recognize the sentiment of the text. This work focuses on the analysis of emotional attitudes in the texts that are the statements of centuries old computer science at the Technical University of Lodz. Due to the free style of expression, the created application is a good basis for automatic analyses under the diagnostic angle, as an aid for psychologists, educators or sociologists. The bag-of-words and n-gram methods were used to vectorize the text, while for the classification of sentiments, k-nn and NBC were used.