A set of tools, and associated web interface, for generating and visualizing automated meta-analyses of functional magnetic resonance imaging (fMRI) data. Uses text mining and machine learning techniques to extract and meta-analyze activation coordinates from published neuroimaging articles. Parts of the codebase are currently available on GitHub. Data are available from neurosynth.org. For details, see the FAQ or the Nature Methods manuscript.
A small Windows application to generate measures of orthographic typicality based on the Levenshtein distance. Given an arbitrary input lexicon, the score for each word reflects the mean distance from that word to the N orthographically closest other words in the lexicon. A slightly more detailed explanation can be found in the README file contained in the .zip archive. Further details and a comparison with a standard measure of orthographic neighorhood size (Coltheart's N) are in a recent paper (Yarkoni, Balota, & Yap, 2008). Note: you probably shouldn't use this program any more; there are now much faster alternatives with greater functionality. See Emmanuel Keuleers' vwr package for R, or Athanassios Protopapas' program in C.
Orthographic and phonological "Levenshtein distance 20" scores for words in the English Lexicon Project. OLD20 reflects the mean distance (in number of steps) from each word to the 20 closest Levenshtein neighbors in the lexicon. PLD20 is the phonological analog, and is computed on the Unisyn representation of ELP words. These measures account for substantially more variance in behavioral measures of lexical decision and pronunciation speed than existing measures such as Coltheart's N.
A broadband 181-item personality measure that can be used to relatively accurately recapture scores in 8 different existing inventories (e.g., the NEO-PI-R, the HEXACO-PI, the MPQ, etc.). The measure was generated using a genetic algorithm-based technique that can be used to abbreviate or compress many other individual differences measures. For further details, see Yarkoni (2010), or this tutorial.
A Ruby package that uses a semantic space model to quantify associations between different concepts in a text (e.g., to estimate political leaning in TV transcripts). For details, see the CASS website, or Holtzman et al (2011).