K-最临近分类算法论文.doc

PAGE \* MERGEFORMAT1 K-最临近分类算法 K-Nearest Neighbour 课程名称 数据仓库与数据挖掘 　 摘 要 K-最临近分类算法（kNN）是著名的模式识别统计学方法，是最好的文本分类算法之一，在机器学习分类算法中占有相当大的地位，是最简单的机器学习算法之一。 随着当今社会，计算机的普遍使用，出现了连绵不断的文本文件，如何对这些毫无逻辑、毫无层次的文件进行分门别类的整理，做到井井有条，层次鲜明呢？? 文本自动分类就是针对上述情况，采用机器，通过一定的约束条件和一些分类算法，自动的对这些文件进行遍历，从而实现分门别类。这样用机器代替人来“阅读”文章，用机器代替人来“整理”文章，不仅减轻了工作人员的负担，而且大大节省了时间，工作人员可以去做更多有意义的事情。 KNN算法的核心思想是，通过计算每个训练样本到待分类数据的距离，取和待分类数据距离最近的K个训练样本， K个样本中哪个类别的训练样本占多数，则待分类数据就属于哪个类别。本文旨在利用K-最临近分类算法进行数据挖掘。 关键词：kNN算法；文本分类；分类算法；数据挖掘； ABSTRACT KNN?algorithm,?a?famous?statistical?method?of?pattern?recognition,which?is?one?of?the?best?algorithms?for?dealing?with?text?categorization,?is?playing?an?important?role?in?machine?learning?classification?algorithm,?and?it?is?one?of?the?simplest?algorithms?in?machine?learning.? With?todays?society,?the?widespread?use?of?computers,?the?continuous?of?the?text?file,? how?about?these?no?logic,?no?level?of?sort,?classify?files?on?do?in?perfect?order,?hierarchy?and?bright? Text?automatic?classification?is?according?to?the?above?situation,?using?the?machine,?through?a?certain?constraint?condition?and?some?classification?algorithm,?automatic?to?traverse?these?files,?so?as?to?realize?classify.?So?using?machines?instead?of?people?to?read,?to?finish,?replacing?workers?with?machines?not?only?reduce?the?burden?of?the?staff,? and?greatly?saves?time?and?staff?do?more?meaningful?things. Thecore?idea?of?KNN?is?to?be?classified,? the?distance?ofdata?calculated?for?each?training?sample,? and?the?data?from?the?nearest?K?training?samples,? which?category?K?sample?training?samples?inthe?majority,? are?to?be?classified?data?belongs?to?which?category. This article is for the purpose of?data mining using K-Nearest Neighbour. Key words：KNN?algorithm, Text?classification, Classification?algorithms, data mining 1.研究背景及现状 国际权威的学术组织the IEEE International Conference on Data Mining (ICDM) 2006年12月评选出了数据挖掘领域的十大经典算法： C4.5，k-Means， SVM， Apriori,，EM， PageRank， AdaBoost，kNN，Naive Bayes,，and CART。 国外对于文本分类的研究开展较