Thesis (Ph.D.)--Chulalongkorn University, 2012

Currently, searching and utilizing social networking systems are becoming a part of daily life. Bibliographic social bookmarking is an example of academic social networking systems. Researchers realize the importance of searching for papers through bibliographic social bookmarking systems, therefore, the integrated indexing and personalized re-ranking for bibliographic social bookmaking systems were created in this dissertation. Three indexing types were developed as a part of research experiment. The results showed that only tagging information is inadequate to create efficient indexing and user profile. The integrated indexing making up with social tagging, title, and abstract (TTA) was a more suitable combination. This technique can be applied to buliding user profile and personalized re-ranking. In addition, the year of publication factor can further support the information about recent papers and improve ranking. Evaluation of the proposed technique was carried out with user-formulated queries. The result found that the year of publication and personalized re-ranking (PTTAYRank) at 90:10 weighted score provided the best personalized re-ranking and could enhance the performance efficiency of the bibliographic social bookmaking systems.