AI-Driven Sentiment Analysis to Optimize Ad Placements in AVOD Platforms

Abstract

The proliferation of Ad-supported Video on Demand (AVOD) platforms has transformed the digital entertainment landscape, where advertisements are vital for monetization. However, poorly timed or irrelevant ads can lead to viewer dissatisfaction and reduced engagement. This research proposes an AI-driven sentiment analysis system to optimize ad placements by understanding the emotional tone of content and aligning ads with viewers' emotional states. By combining sentiment analysis with machine learning models such as BERT, reinforcement learning (RL), and convolutional neural networks (CNNs), this methodology dynamically adjusts ad placements, ensuring a more engaging and non-intrusive viewing experience. The effectiveness of this approach was evaluated through a series of experiments, including sentiment analysis, engagement scores, and user interaction metrics. Results indicate that the integration of sentiment-driven ad placement improves viewer engagement, ad relevance, and overall user satisfaction. This research provides a scalable solution for AVOD platforms to optimize monetization strategies while enhancing the viewer experience.