This Paper presents the work done for the TREC 2010 entity track. We concentrate on constructing enriched anchor text model by exploiting hierarchical information presented in web pages to retrieve promising pages, and heuristic rules to extract potential candidate entities by zooming in the right section.
This paper presents the work done for the TREC 2010 faceted blog distillation task. As the approach used in TREC 2009, a mixture of language models based on global representation is employed to rank the entire blogs by relevance and facets. The parameters in our approach are adjusted according to the experimental results in TREC 2009. In addition, we make use of the results evaluated in TREC 2009 to train ...
Our goal in participating in the TREC 2009 Entity Track is to study whether QA list technique can help improve accuracy of the entity finding task. Also, we take a looking for homepage finding to identify homepages of an entity by training a maximum entropy classifier and a logistic regression models for three types of entity respectively.
This Paper presents the work done for the TREC 2009 faceted blog distillation task of blog track. In our approach, we use a mixture of language models based on global representation. Our model can be regarded as a combination of topic relevance model and faceted relevance model. By pseudorelevance feedback method, we can estimate the above two models from topic relevance feedback documents and facet relevance feedback documents respectively. Experimental ...
Reports by Author
