Research-based Stock Prediction System For Multi-attribute Group Decision Theory

With the progress of society and economy, there are many multi-attribute group decision-making problems in practice. The multi-attribute group decision-making 'problem is going to solve how to aggregate individual preference into the group preference and how to get the ranking of projects according to the preference of the group. The group decision-making we mentioned here is not only a combination of many decision-makers but also a combination of methods which is called combination decision-making in some papers. In this paper, a new data reduction method-multi-attribute group decision-making is used to solve the dock forecasting. The research in this paper is included five aspects: quantizing the qualitative attributes, formalizing the decision-making matrix, ranking the projects, aggregating the individual preference and reducing the attributes.The research achievements of this dissertation are outlined as follows:Analyzed the existing methods of solving the multi-attribute decision-making problems and summarized four methods.The studying on the theory and method of combination decision-making is almost a blank in decision domain. In this paper, the author studies on this kind of group decision-making.In this thesis, data reduction and the reduction of attributes is discussed and studied.Finally, the rough set reduction technique and the multi-attribute group decision-making technique we mentioned in this paper is compared.