A Weakly Supervised and Deep Learning Method for an Additive Topic Analysis of Large Corpora

Abstract

The collaborative effort of a theory-driven content analysis can benefit significantly from the use of topic analysis methods, which allow researchers to add more categories while developing or testing a theory. Additivity also enables the reuse of previous efforts or the merging of separate research projects, thereby increasing the accessibility of such methods and the ability of the discipline to create shareable content analysis capabilities. This paper proposes a weakly supervised topic analysis method, which combines a low-cost unsupervised method to compile a training-set and supervised deep learning as an additive and accurate text classification method. We test the validity of the method, specifically its additivity, by comparing the results of the method after adding 200 categories to an initial number of 450. We show that the suggested method is a solid starting point for a low-cost and additive solution for a large-scale topic analysis.