An Integrated Term-Based Corpus Query System

An Integrated Term-Based Corpus Query System Irena Spasic� Goran Nenadic� Kostas Manios� Sophia Ananiadou Computer Science Dept. of Computation� Computer Science� Computer Science University of Salford� UMIST�University of Salford� University of Salford I.Spasic@salford.ac.uk G.Nenadic@umist.ac.uk K.Manios @salford.ac.uk S.Ananiadou@salford.ac.uk Abstract In this paper we describe the X-TRACT workbench, which enables efficient term- based querying against a domain-specific literature corpus. Its main aim is to aid domain specialists in locating and extracting new knowledge from scientific literature corpora. Before querying, a corpus is automatically terminologically analysed by the ATRACT system, which performs terminology recognition based on the C/NC- value method enhanced by incorporation of term variation handling. The results of terminology processing are annotated in XML, and the produced XML documents are stored in an XML-native database. All corpus retrieval operations are performed against this database using an XML query language. We illustrate the way in which the X-TRACT workbench can be utilised for knowledge discovery, literature mining and conceptual information extraction. 1� Introduction New scientific discoveries usually result in an abundance of publications verbalising these findings in an attempt to share new knowledge with other scientists. Electronically available texts are continually being created and updated, and, thus, the knowledge represented in such texts is more up-to-date than in any other media. The sheer amount of published papers' makes it difficult for a human to efficiently 1 For example, the MEDLINE database (www.ncbi.nlm.nih.gov/PubMed/) currently contains over 12 million abstracts in the domains of molecular biology, biomedicine and medicine, growing by more than 40.000 abstracts each month. localise the information of interest not only in a collection of documents, but also within a single document. The growing number of electronically available knowledge sources emphasises the importance of developing flexible and efficient tools for automatic knowledge mining. Different literature mining techniques (e.g. (Pustejovsky et al., 2002)) have been developed recently in order to facilitate efficient discovery of knowledge contained in large corpora. The main goal of literature mining is to retrieve knowledge that is "buried" in a text and to present the digested knowledge to users. Its advantage, compared to "manual" knowledge discovery, is based on the ability to systematically process enormous amounts of text. For these reasons, literature and corpus mining aim at helping scientists in collecting, maintaining, interpreting and curating domain- specific information. Apart from digesting knowledge from corpora, there is also a need to facilitate knowledge mining via suitable querying systems, which would allow scientists to locate semantically related information. In this paper we introduce X-TRACT (XML-based Terminology Recognition and Corpus Tools), an integrated literature corpora mining and querying system designed for the domain of molecular biology and biomedicine, where terminology-driven knowledge acquisition and XML-based querying are combined using tag- based information management. X-TRACT is built on top of a terminology management workbench and it incorporates a GUI to access the features of the XQuery language that allow users to formulate and execute complex queries against a collection of XML documents. Our main assumption is that the knowledge encoded in scientific literature …

Content truncated to 500 words