Using SSM for Enhancing Summarization

This paper describes a query-based multi-document summarizer that was built to participate in the summarization task of TAC11. The Similar to last year's system, it relies on a thesaurus extracted from Wikipedia and uses it as its underlying ontology. In addition, the system was updated by including a Sentences Simplification Module (SSM) that is applied in an iterative process in the post-processing stage. SSM also affects how sentences are ranked and chosen to form the summaries. The evaluation results and the performance of the system are provided.

[1]  Mourad Oussalah,et al.  Centroid-based Classification Enhanced with Wikipedia , 2010, 2010 Ninth International Conference on Machine Learning and Applications.

[2]  Mourad Oussalah,et al.  Using features extracted from Wikipedia for the task of Word Sense Disambiguation , 2010, 2010 IEEE 9th International Conference on Cyberntic Intelligent Systems.

[3]  Mourad Oussalah,et al.  A Semantic Summarization System: University of Birmingham at TAC 2008 , 2008, TAC.

[4]  Tat-Seng Chua,et al.  A Public Reference Implementation of the RAP Anaphora Resolution Algorithm , 2004, LREC.