D. A multi-document summarization system for sociology dissertation abstracts design, imple.pdf

A Multi-document Summarization System for Sociology Dissertation Abstracts: Design, Implementation and Evaluation Shiyan Ou, Christopher S.G. Khoo, and Dion H. Goh Division of Information Studies School of Communication Information Nanyang Technological University Singapore, 637718 {p assgkhoo, ashlgoh}@.sg Abstract. The design, implementation and evaluation of a multi-document summarization system for sociology dissertation abstracts are described. The system focuses on extracting variables and their relationships from different documents, integrating the extracted information, and presenting the integrated information using a variable-based framework. Two important summarization steps – information extraction and information integration were evaluated by comparing system-generated output against human-generated output. Results indicate that the system-generated output achieves good precision and recall while extracting important concepts from each document, as well as good clusters of similar concepts from the set of documents. 1 Introduction Multi-document summarization has begun to attract much attention in the last few years [6]. A multi-document summary has several advantages over the single- document summary. It provides an integrated overview of a document set indicating common information across many documents, unique information in each document, and cross-document relationships, and allows users to zoom in for more details on aspects of interest. Our present work aims to develop a method for automatic summarization of sets of sociology dissertation abstracts that may be retrieved by a digital library system or search engine in response to a user query. Recently, many digital libraries have begun to provide online dissertation abstract services, since they contain a wealth of high- quality information by specifying research objectives, research methods and results of dissertation projects. However, a dissertati