Collaboratively Tracking Interests for User Clustering in Streams of Short Texts
- 7 May 2018
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Knowledge and Data Engineering
- Vol. 31 (2) , 257-272
Abstract
In this paper, we aim at tackling the problem of user clustering in the context of their published short text streams. Clustering users by short text streams is more challenging than in the case of long documents associated with them as it is difficult to track users' dynamic interests in streaming sparse data. To obtain better performance, we propose two user collaborative interest tracking models (UCIT) that aim at tracking changes of each user's dynamic topic distributions in collaboration with their followees' dynamic topic distributions, based both on the content of current texts and the previously estimated distributions. UCIT models can be either short-tern or long-term dependency topic models. Short-term dependency model collaboratively tracks users' interests based on users' topic distributions at the previous time period only, whereas long-term dependency model collaboratively tracks users' interests based on users' topic distributions at multiple time periods. We also propose two collapsed Gibbs sampling algorithms for collaboratively inferring users' dynamic interests in our short-term and long-term dependency topic models, respectively. We evaluate our proposed models via a benchmark dataset. Experimental results validate the effectiveness of our UCIT models that integrate both users' and their collaborative interests for user clustering by short text streams.Keywords
Funding Information
- Elsevier Foundation
- UCL Big Data Institute
This publication has 32 references indexed in Scilit:
- Dynamically Modeling Patient's Health State from Electronic Medical Records2015
- User Based Aggregation for Biterm Topic Model2015
- Collaborative personalized Twitter search with topic-language models2014
- Large-scale analysis of individual and task differences in search result page examination strategies2012
- Online multiscale dynamic topic models2010
- Data clustering: 50 years beyond K-meansPattern Recognition Letters, 2010
- Contextual factors for finding similar expertsJournal of the American Society for Information Science and Technology, 2010
- Finding scientific topicsProceedings of the National Academy of Sciences of the United States of America, 2004
- Web usage miningACM SIGKDD Explorations Newsletter, 2000
- The Collapsed Gibbs Sampler in Bayesian Computations with Applications to a Gene Regulation ProblemJournal of the American Statistical Association, 1994