الفهرس | Only 14 pages are availabe for public view |
Abstract Decision making is about selecting the best choice from a range of possible options. Based on the number of the participants in the decision making process, it is classified into individual and group decision making. Group decision making presents a number of challenges that need to be handled by a tool that facilitates communication and deliberation. A number of models and frameworks have been proposed to handle specific types of problems due to their limited features. Therefore, there are problems that require the use of more than one model due to the limitations that an existing single model would impose, such as shortage in communications between participants, disallowing participants from adding new alternatives or criteria, or providing no statistics about the problem. The main contribution of the thesis is to build a robust model that can be used over a wide range of problem scenarios by integrates the key features from the models and frameworks that addressed the issues that related to group decision making. The proposed model makes use of a trust and delegation feature to manage the communications among the participants and handle the differences in their levels of expertise. It applies the analytic hierarchical processes as a decision analysis tool for gathering participants decisions. It also introduces the sentiment analysis feature as a way for exploring and explaining the group decisions during and after the process. The model has a generic and flexible architecture which allows it to adapt to the requirements of a given problem. A simple embodiment is implemented and utilized in a small experiment. The experiment evaluation showed that 65% of participants agreed on the final decision and the process fairness; 76% of participants recommended to use the proposed model in the real political processes in which representatives are elected to make decision on behalf of a group. However, we couldn’t compare these results due to the lack of experimental results for the models and frameworks that we build the proposed model upon them. |