clips / pattern

Web mining module for Python, with tools for scraping, natural language processing, machine learning, network analysis and visualization.
https://github.com/clips/pattern/wiki
BSD 3-Clause "New" or "Revised" License
8.76k stars 1.58k forks source link

Negative values for subjectivity in Dutch sentiment analysis #242

Open Aerylia opened 6 years ago

Aerylia commented 6 years ago

The dutch sentiment file (see nl-sentiment.xml) has words with negative subjectivity, which does not respect the boundary values for subjectivity: [0.0, 1.0]. I did not check for how many cases, but the word "verloren" is an example.

Sentiment files for other languages may have negative subjectivity as well. Since this problem is similar to #125, but for Dutch in stead of French. This should be checked and fixed.