Abstract—Sentence Reduction is a valuable task in the framework of text summarization. In previous works, sentence is reduced by removing redudant words or phrases from original sentence and try to remain important information. In this paper we propose a new method that used Viterbi algorithm for find the most likelihood substring and then concatenate them to generate sentence reduction. Reduced sentence not only remain important information from original sentence, grammatically is ensured. The experimental results shown that, our method better than previous works and closely method that has done by human.
Index Terms—Important word, Likehood substring, Sen- tence reduction, Vietnamese text, Virtebi algorithm.
Ha Nguyen Thi Thu, Head of Computer Science Department, Electric Power University, Hanoi, Vietnam. (Email: firstname.lastname@example.org).
Quynh Nguyen Huu, Dean of Information Technology Faculty, Electric Power University, Hanoi, Vietnam. (Email: email@example.com).
Cite: Ha Nguyen Thi Thu and Quynh Nguyen Huu, "Concatenate the Most Likelihood Substring for Generating Vietnamese Sentence Reduction," International Journal of Engineering and Technology vol. 3, no. 3, pp. 203-207, 2011.