Semantic similarity of tags to enchance collaborative filtering of recommender systems

Ghabayen Ayman S.A. (P56771)

Please use this identifier to cite or link to this item: https://ptsldigital.ukm.my/jspui/handle/123456789/513508

Full metadata record

DC Field	Value	Language
dc.contributor.advisor	Shahrul Azman Mohd Noah, Prof. Dr.	-
dc.contributor.author	Ghabayen Ayman S.A. (P56771)	-
dc.date.accessioned	2023-10-16T04:37:26Z	-
dc.date.available	2023-10-16T04:37:26Z	-
dc.date.issued	2014-04-30	-
dc.identifier.other	ukmvital:79910	-
dc.identifier.uri	https://ptsldigital.ukm.my/jspui/handle/123456789/513508	-
dc.description	Collaborative filtering recommender systems play an important role in overcoming the "information overload" phenomenon by providing users with relevant information based on their preferences. Despite the popularity of collaborative filtering recommender system, it still suffers from certain limitations in relation to "cold-start" users, for example, where often there are insufficient preferences to make recommendations. Moreover, there is the data-sparsity problem, where there is limited user feedback data to identify similarities in users' interests because there is no intersection between users' transactional data, a situation which also leads to degraded recommendation quality. Recent years have also seen a significant growth in social tagging systems which allow users to use their own generated tags to organize, categorize, describe and search digital content on social media sites. The growing popularity of tagging systems is leading to an increasing need for automatic generation of recommended items for users. Much previous research focuses on incorporating recommender techniques in social tagging systems to support the suggestion of suitable tags for annotating related items. This research aims to investigate methods to exploit users contributed data in the form of folksonomies for enriching the limited information sources in collaborative filtering recommender systems toward improving the quality of recommendation. The research explores the latent semantic relation between tags and the semantic expansion between tags to enhance the recommendation quality. In both cases WordNet lexical ontology is used to measure the distance between tags. In order to achieve the aim of this research, we proposed new collaborative filtering approaches called semantic tag collaborative filtering (STCF) and semantic tag expansion collaborative filtering (STECF). The proposed approaches exploit the semantic foundation for users' tags to capture users' semantic preferences in social tagging media to determine semantically nearest neighbours. Hence, items which have been considered by these semantically nearest neighbours are ranked and then generate a Top-N list of recommended items which considered semantically relevant to users' needs. Evaluation was conducted on MovieLens and Bibsonomy datasets. The results were compared with the commonly known baseline approaches the user-based collaborative filtering and the item-based collaborative filtering using different similarity methods such as cosine and overlapping methods. The standard recall, precision and harmonic-means measures are used as the evaluation matrices. The experimental results in both datasets show that our proposed approach successfully enhance the recommendation quality of collaborative filtering recommender systems and outperform the baseline approaches. Furthermore, the results indicate that exploiting semantic tag information exist in social tagging media can improve the quality of item recommendation and mitigate the cold start and data sparsity problems encountered in traditional collaborative filtering recommender systems.,Ph.D.	-
dc.language.iso	eng	-
dc.publisher	UKM, Bangi	-
dc.relation	Faculty of Information Science and Technology / Fakulti Teknologi dan Sains Maklumat	-
dc.rights	UKM	-
dc.subject	Semantic	-
dc.subject	Recommender systems	-
dc.subject	Recommender systems (Information filtering)	-
dc.title	Semantic similarity of tags to enchance collaborative filtering of recommender systems	-
dc.type	Theses	-
dc.format.pages	181	-
dc.identifier.callno	QA76.9.I58G483 2014 3 tesis	-
dc.identifier.barcode	001061; 005305(2021)(PL2)	-
Appears in Collections:	Faculty of Information Science and Technology / Fakulti Teknologi dan Sains Maklumat

Files in This Item:

File	Description	Size	Format
ukmvital_79910+SOURCE1+SOURCE1.0.PDF Restricted Access		1.3 MB	Adobe PDF	View/Open

Show simple item record Recommend this item