资 源 简 介
Text Matrix implements a c++ library and corresponding utilities for text mining.
Cross-platform execution are supported.
FeaturesText in Chinese (GBK,GB2312) and English to matrix converting.Sparse matrix storage.Chinese character based ngram, Chinese word based ngram extraction.Chinese word segmentation.Stop word filtering.Various output file format support (svmlight style,cluto ...).Various feature weight (bool, tf, tdidf, various tfidf variant ).Svm style training and testing file load and save.Chi-square feature selection.Standard C++ implement with cross-platform execution.Evaluation ( precision, recall, Fmeasure, micro-averaged F1 ...).Easy to implement your algorithm with library support.Naive bayes and K-nearest neighbor as demonstration algorithms is supplied.KeySubStringGroup feature extracting.
