This study contributes to the academic literature concerning social tag systems for digital libraries, addressing the identified information gap from the user’s perspective. It defines social tagging tools and tests users’ perceptions about possible practices. Moreover, it evaluates the effect when using social tagging systems in digital libraries, to assess whether such a system enhances the search process, and to identify whether there is any significant relationship between using social tagging systems in digital libraries and user satisfaction. Although developments in the field of social tags have been significant in recent years, there remains an open question regarding their usability, particularly in the context of digital libraries. Therefore, there is a need for further investigation, exploration and evaluation, and so this work contributed to this by exploring the usability of social tagging in digital libraries in terms of accuracy for research, user satisfaction and adoptability. For this study, Saudi students were given the opportunity to use the system in the United Kingdom, and their experiences, and opinions regarding ease of use and adoptability were then analysed to determine if they would assist digital libraries in Saudi Arabia to achieve their educational goals and to ensure user numbers would not decrease. A quantitative approach and a qualitative approach were combined to collect and analyse the data used in this research. The two approaches were conducted in sequential phases. In the first quantitative phase, assessment measures were administrated to Saudi students using library websites while studying in the UK. Data was collected from 175 participants, and statistical analysis was conducted using SPSS. Cross tabulation was also used to describe the numerical data and a chi-square analysis was conducted to determine the relationship between the various study variables. In the follow-up qualitative phase, semi-structured interviews were undertaken with 15 Saudi students, to explore the proposed hypothesis in depth. This data was then thematically analysed. Results concerning the usability of social tagging in digital libraries obtained in western universities cannot be generalised to Saudi Arabian universities, because the context of Saudi Arabia differs culturally and academically (Alsurehi & Al Youbi, 1014). To address this, the study utilised a sample of Saudi Arabian students, who had had the opportunity to experience using social tags while studying abroad, specifically in the United Kingdom. Their experience might potentially be very important and this research could be considered a first attempt to examine the usability of social tags in digital libraries. Since to date few empirical studies have directly addressed the usability issues raised here in Saudi Arabia, this research also offers a contribution in this area. In addition, although this study relates to the Saudi perspective, the findings can also be considered valuable to Arab countries sharing similar cultural and academic traditions.

This thesis describes work on using content to improve recommendation systems. Personalised recommendations help potential buyers filter information and identify products that they might be interested in. Current recommender systems are based mainly on collaborative filtering (CF) methods, which suffer from two main problems: (1) the ramp-up problem, where items that do not have a sufficient amount of meta-data associated with them cannot be recommended; and (2) lack of transparency due to the fact that recommendations produced by the system are not clearly explained. In this thesis we tackle both of these problems. We outline a framework for generating more accurate recommendations that are based solely on available textual content or in combination with rating information. In particular, we show how content in the form of social tags can help improve recommendations in the book and movie domains. We address the ramp-up problem and show how in cases where they do not exist, social tags can be automatically predicted from available textual content, such as the full texts of books. We evaluate our methods using two sets of data that differ in product type and size. Finally we show how once products are selected to be recommended, social tags can be used to explain the recommendations. We conduct a web-based study to evaluate different styles of explanations and demonstrate how tag-based explanations outperform a common CF-based explanation and how a textual review-like explanation yields the best results in helping users predict how much they will like the recommended items.

In this thesis, a tag-based approach is proposed for predicting mashup patterns, thus deriving inspiration for potential new mashups from the community's consensus. The proposed approach applies association rule mining techniques to discover relationships between APIs and mashups based on their annotated tags. The importance of the mined relationships is advocated as a valuable source for recommending mashup candidates while mitigating common problems in recommender systems. The proposed methodology is evaluated through experimentation using a real-life dataset. Results show that the proposed mining approach achieves prediction accuracy with 60% precision and 79% recall improvement over a direct string matching approach that lacks the mining information. / Master of Science

In the field of Music Data Mining, Mood and Topic information has been considered as a high level metadata. The extraction of mood and topic information is difficult but is regarded as very valuable. The immense growth of Web 2.0 resulted in Social Tags being a direct interaction with users (humans) and their feedback through tags can help in classification and retrieval of music. One of the major shortcomings of the approaches that have been employed so far is the improper filtering of social tags. This thesis delves into the topic of information extraction from songs’ tags and lyrics. The main focus is on removing all erroneous and unwanted tags with help of other features. The hierarchical clustering method is applied to create clusters of tags. The clusters are based on semantic information any given pair of tags share. The lyrics features are utilized by employing CLOPE clustering method to form lyrics clusters, and Naïve Bayes method to compute probability values that aid in classification process. The outputs from classification are finally used to estimate the accuracy of a tag belonging to the song. The results obtained from the experiments all point towards the success of the method proposed and can be utilized by other research projects in the similar field.

This study explores how the social tags are employed by users of LibraryThing, a popular web 2.0 social networking site for cataloging books, to describe works on Asian women in representing themes within the context of intersectionality. Background literature in the domain of subject description of works has focused on race and gender representation within traditional controlled vocabularies such as the Library of Congress Subject Headings (LCSH). This study explores themes related to intersectionality in order to analyze how users construct meaning in their social tags. The collection of works used to search for social tags came from the Association of College and Research Libraries’ list on East Asian, South and Southeast Asian, and Middle Eastern women. A pilot study was conducted comprising of a limited sample in each of the three domains, which helped generate a framework of analysis that was used in application for the larger sample of works on Asian women. The full study analyzed 1231 social tags collected from 122 works on Asian women. Findings from this study showed that users construct a variety of intersections relating to gender and ethnicity for works on Asian women. Overall findings from this showed that gender and gender-related constructs were the most common subject of tags employed for works on Asian women. Users more often referred to geography rather than ethnicity when describing the materials on Asian women. Interesting themes to emerge involved how gender and other constructs differed among the three domains. Tags describing the majority of East Asia, such as Chinese and Japanese were most common in the East Asian dataset. Countries not considered the “majority” in South and Southeast Asia were often used, such as Indonesia and the Philippines. Themes of sexuality and religion were much more prevalent in the Middle Eastern set of tags. Social tags act as a mechanism for social commentary. Researchers have access to a plethora of constructions available to them through these social tags; such abundance of information is a valuable resource to understanding how the general populace understands intersections and constructs identity.