TextAnalysis
Revision as of 05:14, 24 February 2016 by Gpalermo (talk | contribs) (→Resources for Exploring Text Analysis)
Resources for Exploring Text Analysis
- "Where to Start," courtesy of Ted Underwood
- Stanford's Introduction, from Tooling Up for Digital Humanities
- Brendan O’Connor, et al., "Computational Text Analysis for Social Science"
- Paul Baker, Using Corpora in Discourse Analysis (soon to be available via the NEU Library); covers Corpus Building, Frequency and Dispersion, Concordance, and Collocation
Python
- Folgert Karsdorp, Python Programming for the Humanities
- Charles Severance, Python for Informatics, an applied but comprehensive introductory Python text with sections on text parsing
- Download and install Python
- Download and install PyCharm, an Integrated Development Environment (IDE) for Python
R
- Matthew Jockers, Text Analysis With R for Students of Literature (PDF available for download via the NEU Library)
- Download and install R
- Download and install RStudio, an Integrated Development Environment (IDE) for R
- RSeek, a search tool for finding resources on R
- Simple data types in R
Topic Modeling
- Megan R. Brett's "Basic Introduction" (conceptual)
- Scott Weingart's "Guided Tour" (comprehensive, lots of links)
- Ben Schmidt's article about Latent Dirichlet allocation's (LDA's) limitations
Tools
- MALLET, an open-source and Java-based Latent Dirichlet allocation (LDA) package
- Ben Schmidt's R package wrapping MALLET
- GUI Tools that use MALLET
- Stanford Topic Modeling Toolbox (an alternative to MALLET)
word2vec
- Ben Schmidt's Blog Post on Vector Space Models
- which links to his R wrapper package for word2vec
Miscellaneous text analysis tools
- Voyant Tools
- Laurence Anthony's AntConc, a GUI concordancing and text analysis toolkit
- CasualConc, a Mac OSX-native toolkit (AntConc's Mac version is ported)
- David McClure's TextPlot, a Python package that produces force-directed network of words in a text, the nodes of which are clustered using estimated kernel densities
- Bookworm
Corpus building
Some places to get text
Plain text
- Project Gutenberg
- Early English Books Online (EEBO) (some texts TEI-encoded)
- Early Caribbean Digital Archive (ECDA)