CAJ | 학술논문

企业知识库数据量以爆炸性的速度增长,其中大部分信息是非结构化的文本数据,系统往往不能快速准确地满足企业用户的查询请求.为解决这一问题,提出了一种基于TFIDF算法同义替换和相邻合并的文本挖掘技术.这种技术可以降低服务器压力,使服务人员可以更快更准确的从知识库中寻找出相关信息.最后以实例验证了本算法的有效性.
기업지식고수거량이폭작성적속도증장,기중대부분신식시비결구화적문본수거,계통왕왕불능쾌속준학지만족기업용호적사순청구.위해결저일문제,제출료일충기우TFIDF산법동의체환화상린합병적문본알굴기술.저충기술가이강저복무기압력,사복무인원가이경쾌경준학적종지식고중심조출상관신식.최후이실례험증료본산법적유효성.
Enterprise knowledge repository increases with the explosive growth rate, most of which is unstructured text data.The application often can not meet the user's query requests quickly and accurately. To solve this problem, a novel text mining technique based on TFIDF with synonymous substitutions and adjacent integrationis proposed, which can reduce the server stress and make it more efficient to extract expected information. Finally, examples demonstrate the effectiveness of the algorithm.