Alert button

Federated Neural Topic Models

Add code
Alert button
Dec 05, 2022
Figure 1 for Federated Neural Topic Models
Figure 2 for Federated Neural Topic Models
Figure 3 for Federated Neural Topic Models

Share this with someone who'll enjoy it:

Over the last years, topic modeling has emerged as a powerful technique for organizing and summarizing big collections of documents or searching for particular patterns in them. However, privacy concerns arise when cross-analyzing data from different sources is required. Federated topic modeling solves this issue by allowing multiple parties to jointly train a topic model without sharing their data. While several federated approximations of classical topic models do exist, no research has been carried out on their application for neural topic models. To fill this gap, we propose and analyze a federated implementation based on state-of-the-art neural topic modeling implementations, showing its benefits when there is a diversity of topics across the nodes' documents and the need to build a joint model. Our approach is by construction theoretically and in practice equivalent to a centralized approach but preserves the privacy of the nodes.

* 5 pages, 3 figures  

Share this with someone who'll enjoy it: