The role of named entities in Web People Search

Proceedings of the 2009 Conference on Empirical Methods in Natural Language Processing, pages 534–542, Singapore, 6-7 August 2009. c⃝2009 ACL and AFNLP The role of named entities in Web People Search Javier Artiles UNED NLP & IR group Madrid, Spain javart@bec.uned.es Enrique Amig´o UNED NLP & IR group Madrid, Spain enrique@lsi.uned.es Julio Gonzalo UNED NLP & IR group Madrid, Spain julio@lsi.uned.es Abstract The ambiguity of person names in the Web has become a new area of interest for NLP researchers. This challenging problem has been formulated as the task of clustering Web search results (returned in response to a person name query) according to the individual they mention. In this paper we compare the coverage, reliability and in- dependence of a number of features that are potential information sources for this clustering task, paying special attention to the role of named entities in the texts to be clustered. Although named entities are used in most approaches, our results show that, independently of the Machine Learn- ing or Clustering algorithm used, named entity recognition and classification per se only make a small contribution to solve the problem. 1 Introduction Searching the Web for names of people is a highly ambiguous task, because a single name tends to be shared by many people. This ambiguity has recently become an active research topic and, si- multaneously, in a relevant application domain for web search services: Zoominfo.com, Spock.com, 123people.com are examples of sites which per- form web people search, although with limited disambiguation capabilities. A study of the query log of the AllTheWeb and Altavista search sites gives an idea of the relevance of the people search task: 11-17% of the queries were composed of a person name with additional terms and 4% were identified as person names (Spink et al., 2004). According to the data avail- able from 1990 U.S. Census Bureau, only 90,000 different names are shared by 100 million people (Artiles et al., 2005). As the amount of informa- tion in the WWW grows, more of these people are mentioned in different web pages. Therefore, a query for a common name in the Web will usually produce a list of results where different people are mentioned. This situation leaves to the user the task of find- ing the pages relevant to the particular person he is interested in. The user might refine the original query with additional terms, but this risks exclud- ing relevant documents in the process. In some cases, the existence of a predominant person (such as a celebrity or a historical figure) makes it likely to dominate the ranking of search results, compli- cating the task of finding information about other people sharing her name. The Web People Search task, as defined in the first WePS evaluation cam- paign (Artiles et al., 2007), consists of grouping search results for a given name according to the different people that share it. Our goal in this paper is to study which doc- ument features can contribute to this task, and …

Content truncated to 500 words