Skip to content
Show report in:

UMINF 24.04

Deep Learning for News Topic Identification in Limited Supervision and Unsupervised Settings

In today's world, following news is crucial for decision-making and staying informed. With the growing volume of daily news, automated processing is essential for timely insights and in aiding individuals and corporations in navigating the complexities of the modern world. Another use of automated processing is contextual advertising, which addresses privacy concerns associated with cookie-based advertising by placing ads solely based on web page content, without tracking users or their online behavior. Therefore, accurately determining and categorizing page content is crucial for effective ad placements. The News media, heavily reliant on advertising to sustain operations, represent a substantial market for contextual advertising strategies.

Inspired by these practical applications and the advancements in deep learning over the past decade, this thesis mainly focuses on using deep learning for categorizing news articles into topics of varying granularity. Considering the dynamic nature of these applications and the limited availability of relevant labeled datasets for training models, the thesis emphasizes developing methods that can be trained effectively using unlabeled or partially labeled data. It proposes semi-supervised text classification models for categorizing datasets into predefined coarse-grained topics, where only a few labeled examples exist for each topic, while the majority of the dataset remains unlabeled. Furthermore, to better explore coarse-grained topics within news archives and streams and overcome the limitations of predefined topics in text classification the thesis suggests deep clustering approaches that can be trained in unsupervised settings. Moreover, to address the identification of fine-grained topics, the thesis introduces a novel story discovery model for monitoring event-based topics in multi-source news streams. Given that online news reporting often incorporates diverse modalities like text, images, video, and audio to convey information, the thesis also investigates the synergy between textual and visual elements in news article analysis. To achieve this objective, a text-image dataset was annotated, and a baseline was established for event-topic discovery in multimodal news streams. While primarily intended for news monitoring and contextual advertising, the proposed models could be regarded as novel approaches in semi-supervised text classification, deep clustering, and news story discovery. Comparison with state-of-the-art baseline models demonstrates their effectiveness in addressing the respective objectives.


No keywords specified


Arezoo Hatefi

Back Edit this report
Entry responsible: Arezoo Hatefi

Page Responsible: Frank Drewes