NUS at DUC 2005: Understanding Documents via Concept Links
نویسندگان
چکیده
The primary goal of our participation in DUC 2005 is two-fold. One is to benchmark the performance of a method of computing sentence semantic similarity. The other is to test the effectiveness of a new redundancy minimization formula inspired by Maximal Marginal Relevance(MMR). By using only these two features and eschewing other heuristics, our system performed competitively, achieving the top automated ROUGE scores among participants this year.
منابع مشابه
NUS at DUC 2006: Document Concept Lattice for Summarization
Concepts composed of open-class terms after semantic equivalence discovery can be considered as simplified basic elements. We utilize frequent concept sets to construct a Document Concept Lattice, which contains hierarchical summary information of a document cluster. Based on this lattice, we further extract a set of sentences with maximal representative power and minimal redundancy for summari...
متن کاملEvolving XML Summarization Strategies in DUC
In the Document Understanding Conference for 2005, CL Research made some improvements in its summarization routines based on the use of massively XML-tagged documents containing increasingly richer characterizations of texts. We extended the Knowledge Management System to include an improved capability for identifying redundancy when adding extracted sentences to the summary and for improving t...
متن کاملDUC 2005: Evaluation Of Question-Focused Summarization Systems
The Document Understanding Conference (DUC) 2005 evaluation had a single useroriented, question-focused summarization task, which was to synthesize from a set of 25-50 documents a well-organized, fluent answer to a complex question. The evaluation shows that the best summarization systems have difficulty extracting relevant sentences in response to complex questions (as opposed to representativ...
متن کاملNUS at DUC 2007: Using Evolutionary Models of Text
This paper presents our new, querybased multi-document summarization system used in DUC 2007. Current graph-based approaches to text summarization, such as TextRank and LexRank, assume a static graphmodel which does not model how input text emerges. A suitable evolutionary graph model that is related to human writing/reading process may impart a better understanding of the text and improve the ...
متن کاملCL Research Summarization in DUC 2006: An Easier Task, An Easier Method?
In the Document Understanding Conference (DUC) for 2006, CL Research made a basic change in the method for assessing the significance of sentences in its Knowledge Management Systems summarization routines. This change led to an apparently significant improvement in scores compared to results for DUC 2005, with ROUGE-1 increasing from 0.348 to 0.388. After further detailed comparisons of the DU...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2005