Recommending user connections by utilising the real-time Web
|Title:||Recommending user connections by utilising the real-time Web||Authors:||Hannon, John||Advisor:||Smyth, Barry||Permanent link:||http://hdl.handle.net/10197/6789||Date:||2014||Online since:||2015-08-13T08:45:32Z||Abstract:||Social media services, such as Facebook and Twitter, thrive on user engagement around the active sharing and passive consumption of content. Many of these services have become an important way to discover relevant and interesting information in a timely manner. But to make the most of this aspect of these services it is important that users can locate and follow the most useful producers of relevant content. As these services have continued to grow rapidly this has become more and more of a challenge, especially for new users. This problem can be solved in principle by constructing a recommendation system based on a model of users' preferences and interests to recommend new users worth following.In this thesis we propose a recommendation framework for friend finding. It is capable of integrating different sources of user preference information that is available through services such as Twitter and related services. It is also designed to provide a natural partitioning of user interests based on those topics that are core to the user versus those that are more peripheral and the social connections linked with the user. This provides access to a range of different types of recommendation strategies that may be more helpful in focusing the search for relevant users according to different types of user interests. We demonstrate the effectiveness of our approach by evaluating recommendation quality across large sets of real-world users.||Type of material:||Doctoral Thesis||Publisher:||University College Dublin. School of Computer Science and Informatics||Qualification Name:||Ph.D.||Copyright (published version):||2014 the author||Keywords:||Friend Finding; Recommendation; Social Web; Twitter; User Modelling||Other versions:||http://dissertations.umi.com/ucd:10028||Language:||en||Status of Item:||Peer reviewed|
|Appears in Collections:||Computer Science Theses|
Show full item record
This item is available under the Attribution-NonCommercial-NoDerivs 3.0 Ireland. No item may be reproduced for commercial purposes. For other possible restrictions on use please refer to the publisher's URL where this is made available, or to notes contained in the item itself. Other terms may apply.