quanteda

Information

• Documentation: https://kbenoit.github.io/quanteda/
• Github organization: http://github.org/kbenoit/quanteda
• Docathon project: https://github.com/kbenoit/quanteda/projects/3

Description

quanteda is a fast, flexible toolset for for the management, processing, and quantitative analysis of textual data in R. It includes functions for exploring text, tokenizing texts, managing corpora and associated meta-data, creating document-feature matrixes, computing a variety of text-based statistics, plotting textual representations, computing statistical and machine learning models on texts, applying dictionaries to texts, detecting collocations, and more.

Open Doc issues

• Create standard texts and a dictionary for examples
• create a cheatsheet
• Improve quanteda.io website
• How to left join docvars with those in an existing corpus
• inconsistent output between topfeatures and docfreq
• Wishlist: more documentation / vignettes for textmodels
• topfeatures per-document option
• Document C++ functions, remove those unneeded
• docfreq man page needs clarifying equation(s)
• Complete Jockers vignette

Activity