Sample datasets in arff file format. *[login to view URL] includes document strings and class information. *[login to view URL] files contains tokenized to terms version. Each term or word corresponds to an attribute. These attributes are numeric and shows the frequency of this particular term in a document (instance).
1. Mininews dataset consist of newsgroups postings in 20 different class, each containing 100 postings. A total of 2000 documents (instances).
2. 1150haber dataset consist of news in 5 different classes; each containing 230 news. A total of 1150 documents (instances)
Hi, i have completed masters in Artificial Intelligence. I have good knowledge and expertise of data mining tools and softwares to import the .arff file after transformations in order to apply k nearest neighborhood.