Research Guides: Text Analysis: Tools for Modeling Frequencies, Significant Terms, and Topics

Text Analysis

Tools for Modeling Frequencies, Significant Terms, and Topics

With these tools, you can measure:

How often a specific term/phrase is mentioned
Other terms/phrases often used in conjunction with your terms/phrase
Whether term/phrase usage changed within a specific timeframe
Geographic distribution of authors who use a specific term/phrase

Constellate
Constellate is an online text analysis platform creating by JSTOR Labs. Users input a search term. Optional filters include publication timeframe, format, and language. Constellate pulls from JSTOR's digital library to produce visualizations about term frequency over time.
HathiTrust Digital Library
HathiTrust Research Center Analytics pulls from their vast digital library. The online toolkit includes a Name Entity Recognizer, word cloud, and script-builder to automate local searches in their database.
Orange Text Mining
Orange is a free, downloadable software package with powerful analytical capabilities. Users can explore term frequency, geographic distribution, relationships between terms, as well as analyzing sentiment (e.g. positive vs negative voice). Users have the option to upload their own text data or pull from sources like Wikipedia, Twitter, The Guardian, and the New York Times. Check out this comprehensive set of Orange tutorials.
Palladio
Palladio is an online tool designed Humanities + Design at Stanford University. Users can upload textual datasets and visualize relationships over time and space. This software was specifically designed with Humanities datasets in mind, and is often used to visualize networks captured in letter correspondence.
Sci2Tool
Sci2Tool is downloadable software developed by Indiana University. Users can analyze text and produce a number of visualizations, including maps, word bursts, and networks. This software was produced with an NSF grant and has a huge library of tutorials and scholarly publications.
Scopus
Scopus is an online tool to Constellate. Users input search terms and set parameters. Scopus shows publications within those parameters, and provides analytical tools to explore term frequency over time, author institution, etc. Scopus pulls from the Elsevier database.
sentiment viz
Sentiment Viz is an online tool created by Christopher Healey and Siddarth Ramaswamy. Users input a search term and the tool pulls directly from Twitter. Tweets are visualized by sentiment (e.g. ranging from pleasant to unpleasant), topic and term frequency, as well as author location. The tool is very user friendly. In-depth descriptions about tools and computation processes are available here.
Voyant
Voyant is an online tool for analyzing term frequency across one or more texts. Users provide their own dataset. Visualizations can be shared as a link, image, or embedded html.