Powered Dirichlet-Hawkes Process - Challenging Textual Clustering using a Flexible Temporal Prior
The appearance of a document (news, tweet, article, etc.) is conditioned by its semantic content, but also by its publication date wrt previous articles' ones, to a certain extent. By varying r, we can choose whether to focus our clustering algorithm more on the temporal side or more on the semantic side, or a mixture of both. Here we present wordclouds, triggering kernels and real-intensities for one of the clusters (here related to 2019 Sri Lanka bombings) for various values of r. We say that our model can automatically recover a _story_ (the sequence of related publications and intertwinned their dynamics).

Powered Dirichlet-Hawkes Process - Challenging Textual Clustering using a Flexible Temporal Prior

2021, Dec 07    

Link to the paper

The textual content of a document and its publication date are intertwined. For example, the publication of a news article on a topic is influenced by previous publications on similar issues, according to underlying temporal dynamics. However, it can be challenging to retrieve meaningful information when textual information conveys little information or when temporal dynamics are hard to unveil. Furthermore, the textual content of a document is not always linked to its temporal dynamics. We develop a flexible method to create clusters of textual documents according to both their content and publication time, the Powered Dirichlet-Hawkes process (PDHP). We show PDHP yields significantly better results than state-of-the-art models when temporal information or textual content is weakly informative. The PDHP also alleviates the hypothesis that textual content and temporal dynamics are always perfectly correlated. PDHP allows retrieving textual clusters, temporal clusters, or a mixture of both with high accuracy when they are not. We demonstrate that PDHP generalizes previous work –such as the Dirichlet-Hawkes process (DHP) and Uniform process (UP). Finally, we illustrate the changes induced by PDHP over DHP and UP in a real-world application using Reddit data.

Reference:

Powered Dirichlet-Hawkes Process - Challenging Textual Clustering using a Flexible Temporal Prior
G. Poux-Médard, J. Velcin, S. Loudcher, ICDM, 2021

DOI: 10.1109/ICDM51629.2021.00062