Abstract

This paper presents a proposal for a Decision Support System sensitive to the user’s context in the area of investment. This area is especially complicated due to the complex nature of the stock market. Therefore, a context-sensitive decision support can be a great support for investors. In the literature survey on DSS for investments in the stock market could be found that very little has been explored regarding the investor profile in financial decisions-making systems. Any practical experiment was not found where the investor profile has been applied on the recommendations for investment in the stock market. The work emphasized the main points to be considered in the User Context implementation for decision support systems development. The main motivation for this work was to demonstrate how the performance of Decision Support Systems for investment in stock market could be improved through the application of user context to their recommendation models. A recommendation system for buying and selling of stocks, based on genetic algorithms, was implemented and measured the performance in various test scenarios, with user profiles and without user profile features. The system configured without user profile, often performed below results than the different profiles modeled and implemented. To confirm the preliminary results, the ANOVA test was conducted and the null hypothesis was refuted at 0.0001 level.