TongKey at Entity Track TREC 2011: Related Entity Finding
نویسندگان
چکیده
This paper presents our work done for the TREC 2011 Entity track. A retrieval model was proposed for the task of related entity finding. This model consists of several parts: In order to get more accurate document collection, query analysis method was utilized to format the narrative of each query. Then, our dataset was generated by using ClueWeb09 API 2 . Moreover, we employed the NER tools and text pattern recognition to extract entities from this processed dataset. In particular, the types of target entities are not so well-defined as last year. Therefore, a specific classifier trained by employing Wikipedia titles and category was utilized to identify the categories of target entities. To find related entity homepages and supporting documents, a set of feature-based methods were applied.
منابع مشابه
PRIS at TREC 2011 Entity Track: Related Entity Finding and Entity List Completion
The group of PRIS focuses on both tasks in Entity Track this year, Related Entity Finding (REF) and Entity List Completion (ELC). This paper reports the approaches to the two tasks. In REF, three points are improved based the method of last two year: building entity lexicons including more information, introducing a distance algorithm between keywords and entities to entity ranking, allocating ...
متن کاملLIA-iSmart at the TREC 2011 Entity Track: Entity List Completion Using Contextual Unsupervised Scores for Candidate Entities Ranking
This paper describes our participation in the Entity List Completion (ELC) task at Entity track 2011. Our approach combined the work done for the Related Entity Finding 2010 task with some new criteria as the proximity or the similarity between a candidate answer with the correct answers given as examples or their cooccurrences.
متن کاملLADS: Rapid Development of a Learning-To-Rank Based Related Entity Finding System using Open Advancement
In this paper, we present our system called LADS, tailored to work on the TREC Entity Track Task of Related Entity Finding. The LADS system consists of four key components: document retrieval, entity extraction, feature extraction and entity ranking. We adopt the open advancement framework for the rapid development and use a learning-to-rank approach to rank candidate entities. We also experime...
متن کاملExperiments on Related Entity Finding Track at TREC 2009
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.
متن کامل