Extracting Paraphrases of Japanese Action Word of Sentence Ending Part from Web and Mobile News Articles [chapter]

Hiroshi Nakagawa, Hidetaka Masuda
2005 Lecture Notes in Computer Science  
In this research, we extract paraphrases from Japanese Web news articles that are long and aimed at displaying on personal computer screens and mobile news articles that are short and compact and aimed at mobile terminals' small screens. We have collected them for more than two years, and aligned them at article level and then at sentence level. As the result, we got more than 88,000 pairs of aligned sentences. Next, we extract paraphrases of the final part of sentences from this aligned
more » ... The paraphrases that we try to extract are the sentence final nouns of mobile article sentences and their counterpart expressions of Web article sentences. We extract character strings and word sequencies for paraphrases based on branching factor, frequency and length of string. The precision is 90% for highest ranked candidate and 83% to 59% for each top three candidates of 100 most frequently used action nouns.
doi:10.1007/978-3-540-31871-2_9 fatcat:p4de5alhl5cphmiv5zwgnnxihi