Deep Temporal-Recurrent-Replicated-Softmax for Topical Trends over Time
This work addresses the need for better dynamic topic modeling to analyze temporal trends in unstructured text data, such as research articles, though it appears incremental as it builds on existing neural topic modeling approaches.
The authors tackled the problem of identifying topical trends over time in temporal document collections by introducing RNNRSM, an unsupervised neural dynamic topic model that models latent topical dependencies across time steps. They demonstrated that RNNRSM outperforms state-of-the-art topic models in generalization, topic interpretation, evolution, and trends on 19 years of NLP research articles.
Dynamic topic modeling facilitates the identification of topical trends over time in temporal collections of unstructured documents. We introduce a novel unsupervised neural dynamic topic model named as Recurrent Neural Network-Replicated Softmax Model (RNNRSM), where the discovered topics at each time influence the topic discovery in the subsequent time steps. We account for the temporal ordering of documents by explicitly modeling a joint distribution of latent topical dependencies over time, using distributional estimators with temporal recurrent connections. Applying RNN-RSM to 19 years of articles on NLP research, we demonstrate that compared to state-of-the art topic models, RNNRSM shows better generalization, topic interpretation, evolution and trends. We also introduce a metric (named as SPAN) to quantify the capability of dynamic topic model to capture word evolution in topics over time.