CAJ | 학술논문

针对检索引擎返回的查询结果数量巨大，用户难以在较短的时间获取有用的信息，传统FCM聚类算法的距离无法完全准确描述文本间的相似程度，本文提出了一种新的文本间的距离定义，改进了模糊C均值聚类（FCM）聚类算法，利用新改进的算法对搜索引擎返回的结果进行聚类，方便了用户查询。实验证明了算法的可行性与有效性。
침대검색인경반회적사순결과수량거대，용호난이재교단적시간획취유용적신식，전통FCM취류산법적거리무법완전준학묘술문본간적상사정도，본문제출료일충신적문본간적거리정의，개진료모호C균치취류（FCM）취류산법，이용신개진적산법대수색인경반회적결과진행취류，방편료용호사순。실험증명료산법적가행성여유효성。
The search engine returns large quantity of query results, so it is diffi cult for users to obtain useful information in a relatively short period of time. The traditional FCM clustering algorithm can not describe the similarity distance between texts accurately. This paper puts forward a new defi nition of the distance between texts improving FCM clustering algorithm to make clustering analysis of search engine results to improve effi ciency of inquires. Experiments show that the algorithm is feasible and effective.