AUTOMATIC SUMMARIZATION OF NEWS USING WORDNET CONCEPT GRAPHS

One of the main handicaps in research on automatic summarization is the vague semantic comprehension of the source, which is reflected in the poor quali ty of the consequent summaries. Using further knowledge, as that provided by ontologies, to const ruct a complex semantic representation of the text, can considerably alleviate the problem. In this paper, we introduce an ontology-based extra ctive method for summarization. It is based on mapping the text to concepts and representing the d ocument and its sentences as graphs. We have applied our approach to news articles, taking advan tages of free resources such as WordNet. Preliminar y empirical results are presented and pending problem s are identified.