Information system for determining the emotional color of the text using machine learning
Students Name: Didenko Oleksandr Oleksandrovych
Qualification Level: magister
Speciality: Information Systems and Technologies
Institute: Institute of Computer Science and Information Technologies
Mode of Study: full
Academic Year: 2023-2024 н.р.
Language of Defence: ukrainian
Abstract: Feelings can be very subjective. As humans, we use tone, context and language to convey meaning. How we perceive this information depends on our own experience and unconscious prejudices. In addition to the problem of determining the general mood of the text, any sentence created by man can have several layers of undertone. People express their thoughts in complex ways, often using rhetorical means such as sarcasm, irony and implicit content. This can easily mislead not only the computer that analyzes the text, but also other people. There are a number of techniques and sophisticated algorithms used to teach a computer to analyze the tone of a text, which together give extraordinary results. Being able to quickly see people’s attitudes about a topic, from posts in various forums to news articles, means being one step ahead in developing new strategies and planning future plans. People will always express their feelings and emotions, and as long as they do so in text, the analysis of the emotional color of the text will be relevant issue. This paper describes the prospects of development of use of artificial intelligence for carrying out various analyzes of the text on the Internet and systems-analogues, their features, advantages and lacks were described. The system analysis is also presented in the form of a Tree of Objectives, Hierarchy of Tasks and diagrams. The purpose of this presentation is to demonstrate how each process converts its input data to output and to identify the links between these processes [1]. After that, the methods and means of implementing the system were determined and a user manual was created, which described the main functions, characteristics and features of the program, shows the classes of problem solving. After that, an analysis of a control work example of the developed system was demonstrated. Finally, an assessment of the impact of external and internal factors and the calculation of costs for information system software development was performed. A system analysis of the information system was conducted and the analysis of the research object and subject area was described. The goal tree is defined, at its construction of the goal tree 6 basic criteria of quality of system are also defined. The method analytic hierarchy process (AHP) was used to assess and determine the type of information system, which resulted in the definition of information and analytical system. The statement and substantiation of the problem are described. The conceptual model of the system, the diagram of variants of use, classes, sequence, transitions of states, activity and components are developed. The information system for determining the mood of the text is presented in the form of a website. The modules, database, microservices, client and server part of the web resource have been created. Tools and methods for software implementation are used and the algorithm for using the created functionality is displayed, the sequence of user actions is described and graphically displayed. Using the system, you can determine the general tone of the entered text. The object of research is the processes of determining the emotionality of the text. The subject of research is the methodology of analysis of the emotionality of the text by its semantic component. The aim of the study. The purpose of the bachelor’s thesis is to create an information system for determining the emotionality of the text using machine learning. Research results. Information system gives users the opportunity to get the result of text analysis by its emotional component, save these results in history or on a local device and view them in the future. Key words: text analysis, mood analysis, artificial machine learning, information system, information technologies. List of used literature sources. 1. О. Tomaszewski, H. Tsehelyk, M. Viter, V. Dubuk "Information Technology and Business Process Modeling". Publisher: "Center for Educational Literature" - Kyiv, 2012. - 296 p.