Decoding content taxonomy in video for industry-specific contextual advertising

Tools

De Silva, Waruna and Fernando, Anil (2026) Decoding content taxonomy in video for industry-specific contextual advertising. Multimedia Tools and Applications, 85 (2). 166. ISSN 1380-7501 (https://doi.org/10.1007/s11042-026-21243-4)

[thumbnail of De-Silva-Fernando-MTA-2026-Decoding-content-taxonomy-in-video-for-industry-specific-contextual-advertising]
Preview
 Text. Filename: De-Silva-Fernando-MTA-2026-Decoding-content-taxonomy-in-video-for-industry-specific-contextual-advertising.pdf
Final Published Version
License: Creative Commons Attribution 4.0 logo
Download (2MB)| Preview

Abstract

The exponential growth of free video platforms is redefining content consumption and its advertising dynamics. Beyond this, the progressive irrelevance of cook-ies really underlines the necessity for finding a genuinely new and proper way of recording and analyzing user behavior, with an implication for putting user privacy first by design, compliant with data protection and regulatory guidelines. As a result, contextual advertising has become a suitable strategy for the deliv-ery of relevant and personalized advertisements to users. In light of this evolving process, this article proposes a novel framework to facilitate the discovery of con-textual information on video content within the industrial context. The proposed framework seamlessly integrates the visual and audio features that are extracted from video content to obtain a comprehensive comprehension of videos. This method optimizes the presentation of ads within a context using a combination of multimodal analysis, industry taxonomy, and contextual advertising. The effec-tiveness of the framework is validated through experimental results using the YouTube-8M data set, demonstrating its potential to revolutionize contextual advertising by capturing the essence of video content and aligning it with the industry content taxonomy.

ORCID iDs

De Silva, Waruna and Fernando, Anil ORCID logoORCID: https://orcid.org/0000-0002-2158-2367;
CORE (COnnecting REpositories)