Finding People and Moods
Speaker: Maarten de Rijke
Institution:
ISLA, University of Amsterdam
Abstract:
Much of today's research in retrieval is aimed at going beyond document retrieval: instead of returning a ranked list of documents in response to a query, return ranked lists of specific entities, such as people, or opinions, or answers, or locations, or ...
In the talk I will talk about recent work aimed at finding expert, given a topic, and related work on finding people's moods as they are typically associated with a given topic. Most of this work is uses generative language modeling techniques, while the data for the expert finding task is made available by the W3C, the world wide web consortium and the data for our mood related work is made available by LiveJournal, a popular weblogging platform.
This is joint work with Leif Azzopardi, Krisztian Balog and Gilad Mishne.