BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Towards the Profiling of Twitter Users for Topic-Based Filtering +
  Collaborative Filtering For Recommendation  - Sandra Garcia Esparza + Ste
 ven Bourke  (University College Dublin)
DTSTART:20121210T160000Z
DTEND:20121210T170000Z
UID:TALK41982@talks.cam.ac.uk
CONTACT:Eiko Yoneki
DESCRIPTION:Talk1 - Towards the Profiling of Twitter Users for Topic-Based
  Filtering: Towards the Profiling of Twitter Users for Topic-Based Filteri
 ng: Real-time information streams such as Twitter have become a common way
  for users to discover new information. For most users this means curating
  a set of other users to follow. However\, at the moment the following gra
 nularity of Twitter is restricted to the level of individual users. Our re
 search has highlighted that many following relationships are motivated by 
 a subset of interests that are shared by the users in question.\n\nFor exa
 mple\, user A might follow user B because of their technology related twee
 ts\, but shares little or no interest in their other tweets. As a result t
 his all-or-nothing following relationship can quickly overwhelm users’ t
 imelines with extraneous information. To improve this situation we propose
  a user profiling approach based on the topical categorisation of users’
  posted URLs. These topics can then be used to filter information streams 
 so that they focus on more relevant information from the people they follo
 w\, based on their core interests. In particular\, we have built a system 
 called CatStream that provides for a more fine-grained way to follow users
  on specific topics and filter our timelines accordingly. We present the r
 esults of a live-user study that shows how filtered timelines offer a bett
 er way to organise and filter their information streams.\n\nSandra Garcia 
 Esparza graduated in Computer Science at Universitat Ramon Llull (Barcelon
 a) in 2008. In 2009 she graduated from Trinity College Dublin with a MSc i
 n Networks and Distributed Systems. She is currently in Dublin working on 
 her PhD in CLARITY: Centre for Sensor Web Technologies (University College
  Dublin) in the area of Recommender Systems. Her PhD is about harnessing r
 eal-time data such as Twitter data to provide more personalised user exper
 iences\, including product recommendations and stream filtering.\n\nTalk2 
 - Collaborative Filtering For Recommendation: In Online Social Networks In
  the past recommender systems have relied heavily on the availability of r
 atings data as the raw material for recommendation.\nMoreover\, popular co
 llaborative filtering approaches generate recommendations by drawing on th
 e interests of users who share similar ratings patterns. This is set to ch
 ange because the unbundling of social networks (via open APIs)\, providing
  a richer world of recommendation data. For example\, we now have access t
 o a richer source of ratings and preference data\, across many item types.
  In addition\, we also have access to mature social graphs\, which means w
 e can explore different ways of creating recommendations\, often based on 
 explicit social links and friendships. In this paper we evaluate a convent
 ional collaborative filtering framework in the context of this richer sour
 ce of social data and clarify some important new opportunities for improve
 d recommendation performance.\n\nSteven Bourke is currently a 3rd year PhD
  student working in the area of recommender systems. Previously he complet
 ed a MSc in Computer Science in Trinity College Dublin and a BEng in Softw
 are Engineering in the University of Wales. His research interest lay with
 in the realm of social recommender systems and intelligent user interfaces
  which can make use of social data. He is currently being supervised by Pr
 ofessor Barry Smyth and based in CLARITY: Centre for Sensor Web Technologi
 es\, a research centre in University College Dublin.\n\n
LOCATION:FW26\, Computer Laboratory\, William Gates Builiding
END:VEVENT
END:VCALENDAR