Leveraging ChatGPT for Sponsored Ad Detection and Keyword Extraction in YouTube Videos

2025-02-20

AI Summary

This study addresses the automatic detection and content relevance analysis of spoken sponsorships in YouTube educational videos. We propose the first end-to-end framework integrating prompt engineering, KeyBERT-based keyword extraction, and multi-turn large language model (LLM) reasoning. Our method introduces a novel fine-grained taxonomy comprising nine primary educational content categories and four sponsorship subtypes. Leveraging GPT-4o (with structured prompting and iterative reasoning), KeyBERT, and subtitle text analysis, we achieve high-precision segmentation of sponsorship segments across 421 videos. Experimental results reveal pervasive product-oriented sponsorships within educational content, demonstrating our approach’s superior accuracy, scalability, and cost-efficiency for large-scale auditing. The framework establishes a new paradigm for assessing compliance in online educational content and advancing empirical research on digital advertising ecosystems.

Technology Category

• Natural Language Processing: Fact-Checking / Misinformation Detection (NLP Focus)
• Machine Learning: Large Multimodal Models (LMMs)
• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval

Application Category

• Web Mining and Content Analysis: Large pretrained models with web data
• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work
• Search and Retrieval-Augmented AI: Search Tool Learning with LLM: Teaching LLMs to invoke search and make use of retrieved information

Abstract

This work-in-progress paper presents a novel approach to detecting sponsored advertisement segments in YouTube videos and comparing the advertisement with the main content. Our methodology involves the collection of 421 auto-generated and manual transcripts which are then fed into a prompt-engineered GPT-4o for ad detection, a KeyBERT for keyword extraction, and another iteration of ChatGPT for category identification. The results revealed a significant prevalence of product-related ads across various educational topics, with ad categories refined using GPT-4o into succinct 9 content and 4 advertisement categories. This approach provides a scalable and efficient alternative to traditional ad detection methods while offering new insights into the types and relevance of ads embedded within educational content. This study highlights the potential of LLMs in transforming ad detection processes and improving our understanding of advertisement strategies in digital media.

Problem

• Detect sponsored ads in YouTube videos
• Compare ads with main content
• Extract keywords using advanced AI models

Innovation

• Utilizes GPT-4o for ad detection
• Employs KeyBERT for keyword extraction
• Applies ChatGPT for category identification