Content Word Selection Models for Gist Generation. Short Summary Generation

      Аннотация: Now a days, enormous number of news articles are reported and disseminated on the web. Extracting important information from the news articles to reduce the reading time is the essential issue. Gist generation is an important, difficult and interesting Natural Language Processing (NLP) problem as it requires to mine the essential content words from an article and also to generate a gist that expresses the summary of an article. In ideal case, summary of an article need to generate directly from the understanding of an article. But, developing such type of NLP systems are not possible. This book introduces the importance of the Short Summary Generation. It discusses the various approaches for Content Word Selection from the article and compares the performance of the methods with various measures such as precision, recall and F-measures. It also presents the different preprocessing techniques for improving the performance of the methods.

Ключевые слова: Text Summarization, Naive bayes, Decision Tree, Language Modelling Approach