Loading…

Group Recommendation Systems Based On Pairwise Preference Data And User Contribution

Users’ contributions play a critical role in shaping the trajectory of group decision- making processes, exerting a significant influence on the overall dynamics and out- comes of collaborative efforts. Actively engaging in discussions and sharing their ideas, viewpoints, and preferences, users cont...

Full description

Saved in:
Bibliographic Details
Main Author: Hareendran Nair,Nithya
Format: Dissertation
Language:English
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Users’ contributions play a critical role in shaping the trajectory of group decision- making processes, exerting a significant influence on the overall dynamics and out- comes of collaborative efforts. Actively engaging in discussions and sharing their ideas, viewpoints, and preferences, users contribute diverse perspectives that enrich the collective deliberation and ultimately enhance the effectiveness of decision-making processes. Recognizing the profound impact of individual contributions, this paper introduces a pioneering approach for group recommendation systems grounded in cooperative game theory. Specifically, the proposed methodologies harness the prin- ciples of Wonderful Life Utility (WLU) and Shapley Value (ShV) to quantitatively eval- uate the input of each group member in the decision-making process. By systemat- ically assessing the contributions of users within the group, these techniques aim to foster a more equitable and inclusive decision-making environment, where individual preferences are duly acknowledged and integrated into the collective decision-making process. This innovative approach represents a significant advancement in the field of GRS, offering a nuanced understanding of the role of user contributions and provid- ing a robust framework for facilitating collaborative decision-making in diverse group settings. This method is designed with the primary objective of improving the ac- curacy and fairness of group recommendations within collaborative settings, with a particular emphasis on the significance of individual preferences. To evaluate its ef- fectiveness across a range of group configurations, two distinct datasets—the food dataset and the car dataset—were employed for validation purposes. Random and GcPp clustered groups were formed for each dataset, allowing for a comprehensive assessment of the proposed techniques under different scenarios. The experimental results obtained from these evaluations demonstrate the robustness and efficacy of the system, as it consistently delivers highly accurate and equitable group recommenda- tions across diverse group configurations. These findings highlight the versatility and adaptability of the proposed approach, indicating its potential applicability in a wide array of collaborative contexts. Overall, this study contributes a novel perspective to the field of Group Recommendation Systems (GRS), shedding light on the critical role played by personal preferences in driving meaningf