Keyword extraction method for machine reading comprehension based on natural language processing

Abstract The traditional keyword extraction method has the problem that accuracy of identification and decreases when amount data is increasing. This paper introduces natural language processing technology designs a for machine reading comprehension. In this process, considering information in text comprehension may come from web page, we can refer to NLP system algorithm. Firstly, analyze content single document, then short string level, finally organize preprocessing information, so as achieve Chinese text. At same time, SKEA algorithm designed merge words with meaning, locate location document resources, identify annotated resource base. After annotation completed, extracted. addition, comparative experiment presented prove effectiveness proposed compared method.