Question Prediction Language Model

Question Prediction Language Model Luiz Augusto Pizzato and Diego Moll´a Centre for Language Technology Macquarie University Sydney, Australia {pizzato, diego}@ics.mq.edu.au Abstract This paper proposes the use of a language representation that specifies the relationship between terms of a sentence using question words. The proposed representation is tai- lored to help the search for documents con- taining an answer for a natural language question. This study presents the construc- tion of this language model, the framework where it is used, and its evaluation. 1 Introduction Although Information Retrieval (IR) can be helped by NLP techniques such as named entity (NE) recognition, phrase extraction and syntax parsing (Strzalkowski, 1999), they are not generally used due to their high complexity. One such task that peo- ple can perform somewhat easily whilst still being hard for computers is the answering of factoid ques- tions based on textual content. The Question An- swering (QA) Track of TREC (Voorhees, 2005) fo- cuses on answering questions using the AQUAINT corpus (Graff, 2002), which contains 375 million words from three different sources of newswire data: Xinhua News Service (XIE) from People’s Republic of China, the New York Times News Service (NYT), and the Associated Press Worldstream News Service (APW). For the QA task, not only it is important to find an answer in a document, but also to find the documents that might contain the answer in the first place. Most QA systems take the approach of using off-the-shelf IR systems to return a list of documents that may contain an answer, and then processing the list of documents to look for the required answer. Nor- mally the processing time for every question in these systems is long because of the sheer amount of work that is required after the list of document is returned. Many QA systems focus on the input and output of IR systems. For example, Dumais et al. (2002) perform a passive-to-active voice transformation of the question, in an attempt to bring the IR query closer to the document it is expected to retrieve. Some IR work focuses on improving QA by pas- sage retrieval re-ranking using word overlap mea- sures. For instance, Tellex et al. (2003) compare a group of passage retrieval techniques and conclude that those that apply density-based metrics1 are the most suitable to be used for QA. Some work has been done on IR models that specifically aid the QA task. The work of Monz (2004) defines a weighting scheme that takes into consideration the distance of the query terms. Murdock and Croft (2004) propose a translation lan- guage model that defines the likelihood of the ques- tion being the translation of a certain document. Tiedemann (2005) uses a multi-layer index contain- ing more linguistic oriented information and a ge- netic learning algorithm to determine the best pa- rameters for querying those indexes when applied for the QA task. Tiedemann argues that since ques- tion answering is an all-natural language task, lin- guistic oriented IR will help finding better docu- …

Content truncated to 500 words