Mixtures of Unsupervised Lexicon Classification

Abstract

This paper presents a mixture version of the method-of-moment unsupervised lexicon classification by an incorporation of a Dirichlet process.

A draft on lexicon classification unsupervised learning. It shows that aggregating lexicon scores is equivalent to a finite mixture of multinomial Naive Bayes models. A very preliminary work of a few days man-hours, like a weekly report/note, but might be useful