Emotion Recognition System Project of English Newspapers to Regional E-Business Adaptation

Abstract

Main goal of the work is emotion recognition and sentiment analysis of quotes from English newspapers. Key of this problem is task solution of text emotional analysis, that is, feelings analysis of author who expresses own opinion. Emotional or sentimental text analysis is primarily the classification algorithm aimed at finding certain point and its location and information highlighting particular interest in the process. In this work, the problem of multi-class or binary classification is solved using machine learning. Publication describes general information about human language processing, brief description of analogues, use of machine learning to analyse emotions in text. Article describes the sequence of intellectual system project realization actions and the used tools for emotion recognition. Then, the statistics of tasks performed by neural network with various functions of activators are described. In the publication control example of demo version of the intellectual system project was made, and the main usage scenario was depicted. At the end, the results of the study have been summarized. © 2022 IEEE.