1. UKM Learning and Research Repository
2. Master Thesis / Tesis Master
3. Faculty of Information Science and Technology / Fakulti Teknologi dan Sains Maklumat

Title:	Social tag recommendation using word similarity measure
Authors:	Ali Abdul Raheem Abdul Jabbar (P56157)
Supervisor:	Lailatul Qadri Zakaria, Dr.
Keywords:	Social tag recommendation Information retrieval
Issue Date:	3-Mar-2015
Description:	The Social Web is a set of social relations that connect people through basis of online activities including shopping, education, online games and other social networking applications such as Facebook, Twitter, Flickr and so on. With the expansion of the Social Web and the digital cameras, storage capacities are widening with hundreds of photos shared through these applications. Most of the Social Web applications allow users to describe their photos by using tagging approach. However, since the tagging process is an optional process, most of the photos were left untagged or with insufficient number of tags. It is difficult for information retrieval to search for these photos. In order to overcome this issue, our research aims to develop an automatic tag propagation tool, which will enrich an initial tag with other related tags by using word co-occurrence analysis. The research had used three tag co-occurrence analysis, such as Dice, Cosine and Mutual Information. These analyses enable the tool to identify and recommend related tags based on word similarity measures. The analyses are initiated by collecting images and their tags from FLICKR by using its API. The keywords used were Malaysia, Island, Tourism and Beach. After the tag collection, tag pre-processing analysis was programmed in three steps which are stop word filtering, stemming from using the Porter Stemmer and word normalization to obtain clean tags. Finally, the clean tags were submitted to the word co-occurrence analyzer. Based on the tag iteration analysis, Dice and Cosine provide better tag propagation compared to Mutual Information. Therefore, we have combined the results from both analysis based to support the automatic tag propagation. To sum up, the study concludes that the development the automatic tag propagation has allowed us to recommend relevant tags to support tagging process thus hopefully will be useful to improve information retrieval.,Master/Sarjana
Pages:	73
Call Number:	QA276.4.A238 2015 3 tesis
Publisher:	UKM, Bangi
URI:	https://ptsldigital.ukm.my/jspui/handle/123456789/476232
Appears in Collections:	Faculty of Information Science and Technology / Fakulti Teknologi dan Sains Maklumat

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.