首页>
外国专利>
Topic word acquisition apparatus, method, and program
Topic word acquisition apparatus, method, and program
展开▼
机译:主题词获取装置,方法和程序
展开▼
页面导航
摘要
著录项
相似文献
摘要
PROBLEM TO BE SOLVED: To acquire a topic word which places importance on whether or not the topic word is relevant to specific information represented by at least one of date and place.SOLUTION: A document acquisition unit 12 searches and acquires a document relating to an input keyword and date from a document index 20, and a topic word candidate extraction unit 14 divides the document as the search result into words/characters, generates divided components starting with respective words/characters and ending with the last of the document, rearranging the generated divided components in the order of words/characters, and extracts a topic word candidate on the basis of the number of matching words/characters from heads between adjacent divided components of the rearranged divided components. A date-relevant topic word acquisition unit 16 searches in the document index 20 to obtain the number of documents including both the topic word candidate and the date, the number of documents including only the topic word candidate, the number of documents including only the date, and the number of documents including neither the topic word candidate nor the date and, if a chi-square value calculated by using these numbers is equal to or larger than a threshold, acquires the topic word candidate as a topic word having high relevance to the date.
展开▼