cntext: a Python tool for text mining

Summary

cntext is a text analysis package that provides semantic distance and semantic projection based on word embedding models. Besides,cntext also provides the traditional methods, such as word count , readability, document similarity, sentiment analysis, etc. The link of cntext repo is https://github.com/hidadeng/cntext .

Statement of need

Human society is rich in cognitions, such as perception, thinking, attitude and emotion.As a carrier of ideas, texts not only reflect people's mental activities from the individual’s level but also reflects collective culture from the organizational and social level.In the field of social sciences, the main research path is mining personal mental activities and culture changes in society through text data[@tausczik2010psychological]

There are two common text analysis algorithms: dictionary-based method and word-embedding based method. The cntext library contains both types of algorithms.