lexgen

Documentation

Overview

lexgen converts Chen-Skiena sentiment lexicon text files into an embedded Go source file.

Usage:

go run tools/lexgen/lexgen.go \
  -pos /tmp/pos_clean.txt \
  -neg /tmp/neg_clean.txt \
  -o pkg/analyzers/sentiment/lexicons/lexicon_data.gen.go

The input files use the format: "word lang_code\n" (one per line). Positive words receive valence +1.5, negative words receive -1.5.