Machining technologies to compute the psychological-financial index

Abstract

Subject. This article deals with the issues of measuring the emotions of financial decision-makers based on machine learning technologies in finance. Objectives. The article aims to develop a methodology for measuring emotions in the financial sector, algorithms and software based on artificial intelligence for its implementation in finance. Methods. For the study, we used the methods for marking up text data, index method, machine learning methods, computer technologies for natural language processing, methods for assessing emotions with their own rating scale, induction and deduction, and statistical methods for processing observation results. Results. The article reveals the content of methods for comprehensive measurement of the emotions of financial decision-makers. Software tools were created to implement machine learning technologies when computing an emotion measurement index. Conclusions and Relevance. A psycho-financial index based on automatic calculation using new machine learning technologies provides fundamental analysts of financial markets with an analytical tool that brings the forecast result closer to real conditions, when the emerging combinations of traditional and psychological indicators in the markets provide a new interpretation of current events. This improves the quality of analytical work and improves its results. The result can be used in the work of a fundamental analyst of financial markets. The use of new indicators complements and expands the classic analytical tools and improves the quality of market forecasts.