Real-time decision support and information gathering system for financial domain
| Autor: | Piotr J. Gmytrasiewicz, Chiu-Che Tseng |
|---|---|
| Rok vydání: | 2006 |
| Předmět: |
Statistics and Probability
Decision support system Computer science Decision quality Decision tree Decision field theory Evidential decision theory computer.software_genre Value of information Executive information system Business decision mapping Influence diagram Decision engineering business.industry Evidential reasoning approach Intelligent decision support system Bayesian network Decision rule Condensed Matter Physics R-CAST Management information systems Risk analysis (engineering) Knowledge base Data mining business computer Decision model Optimal decision Decision analysis |
| Zdroj: | Physica A: Statistical Mechanics and its Applications. 363:417-436 |
| ISSN: | 0378-4371 |
| Popis: | The challenge of the investment domain is that a large amount of diverse information can be potentially relevant to an investment decision, and that, frequently, the decisions have to be made in a timely manner. In this setting, an investor has to make decisions considering not only the possible complexity and without full knowledge of the environment, but also the real-time situation of the current market. This presents the potential for better decision support, but poses the challenge of building a decision support agent that gathers information from different sources and incorporates it for timely decision support. These problems motivate us to investigate ways in which the investors can be equipped with a flexible real-time decision support system to be practical in time-critical situations. The flexible real-time decision support system considers a tradeoff between decision quality and computation cost. For this purpose, we propose a system that uses the Object-Oriented Bayesian Knowledge Base (OOBKB) design to create a decision model at the most suitable level of detail to guide the information gathering activities, and to produce an investment recommendation within a reasonable length of time. The more detailed models, located at the bottom of the class hierarchy, are more time consuming to work with and may require more information gathering, but provide better quality of decision support. The more abstract models, located higher in the class hierarchy, can be evaluated faster but result in less precise recommendations. To determine the suitable level of detail we define and use the notion of urgency, or the value of time. Using it, our system can trade off the quality of support the model provides versus the cost of using the model at a particular level of detail. The decision models our system uses are implemented as influence diagrams. Using a suitable influence diagram, our system computes the value of consulting the various information sources available on the web, uses web agents to fetch the most valuable information, and evaluates the influence diagram thereby producing the buy, sell and hold recommendations. We validate our system with experiments in a simplified investment domain. The experiments show that our system produces a quality recommendation under different urgency situations. The contribution of our system is that it provides the flexible decision recommendation for an investor under time constraints in a complex environment. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full Text from ScienceDirect