<i>Wordify:</i> A Tool for Discovering and Differentiating Consumer Vocabularies

Abstract This work describes and illustrates a free easy-to-use online text-analysis tool for understanding how consumer word use varies across contexts. The tool, Wordify, uses randomized logistic regression (RLR) to identify the words that best discriminate texts drawn from different pre-classified corpora, such as posts written by men versus women, or containing mostly negative positive valence. We present illustrative examples show can be used diverse purposes (1) uncovering distinctive vocabularies consumers when writing reviews on smartphones PCs, (2) discovering in Tweets differ between presumed supporters opponents of controversial ad, (3) expanding dictionaries dictionary-based sentiment-measurement tools. empirically Wordify’s RLR algorithm performs better at discriminating than support vector machines chi-square selectors, while offering significant advantages computing time. A discussion is also provided Wordify conjunction with other tools, probabilistic topic modeling sentiment analysis, gain more profound knowledge role language behavior.