cympfh / nlptk

NLP tool kit

nlptk

NLP toolkit written in C++11

For detail, please see wiki or README file in each directory.

ngram

• enumerate ngrams of a document
• allowing multi-document
• outputs pairs of frequency and n-gram

tfidf

• calculates from tf-file

aprf

• caluculate Acc, Prec, Rec and F1-socre
• answer and predict files are written in SVM-light like format

TODO

micro,macro-avg