Here’s a list of a few tools I put together for my work (all written in R):
- TATE: Text Analysis Tools for English is an R package where I wrapped a few functions for text quantification. It uses external norms of valence, arousal, dominance, concreteness, humor, extremity, and emotionality. The functions take a string as input, execute lemmatization of the string, and return a value.
- (the first?) R package for Hebrew sentiment analysis. Provides norms of valence and arousal for up to 1.4M Hebrew words. It was built on E-Millim (Armony-Sivan, Cojocaru, & Babkoff, 2013) and expanded using fasttext (Joulin et al., 2016) and the SVLM Hebrew Wikipedia Corpus (Silber-Varod, Latin, & Moyal, 2017).
- A mini R package to compare pdf and word documents and see how similar they are. Designed to catch homework cheaters and celebrate Pokemon. Co-written with Mattan S. Ben-Shachar. Also available as a Shiny App over here.