Extraction of news topics in quality papers using topic models based on word embeddings

The case for obesity-related topics in the U.K. and the U.S.A.

Abstract

In this paper, we analyze newspaper articles using the topic model, a method for extracting potential topics, to obtain relevant information on the discipline of food systems. As an example, we analyzed obesity-related articles published in the New York Times and The Guardian. Moreover, the topic model used in this study is based on the word embeddings acquired from distributed representation; time transition and topic similarity analyses were performed by this model. In this manner, the effectiveness of the topic model for finding newspaper articles was demonstrated.