Sistema evolucionário de suporte à decisão de investimento em ações usando contexto de usuário

Abstract

There are many difficulties to find the right time to buy and sell shares on the Stock Market. This decision requires analysis of a large amount of information in a short period, also needs discipline and dedication investor. However, the Behavioral Finance, investors in the stock market are less rational than they realize, and often make decisions contrary to their profile. To assist the investment decision process, several systems have been developed for this purpose. This paper presents a review of literature on Decision Support Systems for stock trading, and an analysis of the main models of Finance and Artificial Intelligence techniques used in these systems. Through this research, it was found that this area lacks systems that make investments in accordance with the objectives and preferences of different user profiles recommendations. To contribute to solving this problem, the work proposed the implementation of User Context to improve the recommendations of buying and selling shares of Decision Support Systems for stock trading. To evaluate this hypothesis, the investment system based on Genetic Algorithms of LIPINSKI (2008) was implemented and features of Conservative, Moderate and Aggressive investors profiles was added. The tests compared the system with user profile and without user profile features and demonstrated that it is possible to improve the results of Decision Support Systems for stock trading using User Context. The classes developed to implement user profiles are available for reuse in other systems