Indonesian Journal of Electrical Engineering and Computer Science Vol. 39, No. 1, July 2025, pp. 353~363 ISSN: 2502-4752, DOI: 10.11591/ijeecs.v39.i1.pp353-363  353 Journal homepage: http://ijeecs.iaescore.com Enhancing TV program success prediction using machine learning by integrating people meter audience metrics with digital engagement metrics Khalid El Fayq, Said Tkatek, Lahcen Idouglid Laboratory of Computer Sciences Research, Faculty of Science, Ibn Tofail University, Kenitra, Morocco Article Info ABSTRACT Article history: Received Aug 27, 2024 Revised Mar 11, 2025 Accepted Mar 25, 2025 With the emergence of numerous media services on the internet, television (TV) remains a highly demanded medium in terms of reliability and innovation, despite intense competition that compels us to devise strategies for maintaining audience engagement. A key indicator of a TV channel’s success is its reach, representing the percentage of the target audience that views the broadcasts. To aid TV channel managers, the industry is exploring new methods to predict TV reach with greater accuracy. This paper investigates the potential of advanced machine learning models in predicting TV program success by integrating people meter audience metrics with digital engagement metrics. Our approach combines convolutional neural networks (CNNs) for processing digital engagement data, long short-term memory (LSTM) networks for capturing temporal dependencies, and gaussian processes (GPs) for modeling uncertainties. Our results demonstrate that the best-performing hybrid model achieves a prediction accuracy of 95%. This study contributes to the field by addressing manual scheduling errors, financial losses, and decreased viewership, providing a more comprehensive understanding of audience behavior and enhancing predictive accuracy through the integration of diverse data sources and advanced machine learning techniques. Keywords: Audience reaches Data regression models Digital engagement Media analytics Television audience prediction Watermarking audience This is an open access article under the CC BY-SA license. Corresponding Author: Khalid El Fayq Laboratory of Computer Sciences Research, Faculty of Science, Ibn Tofail University Kenitra, Morocco Email: khalidelfayq@gmail.com 1. INTRODUCTION The marketing industry is one of the largest globally, with television (TV) companies investing millions of dollars in advertising. Audience ratings play a pivotal role in guiding advertisers to tailor campaigns and helping content creators and TV networks refine programming strategies to maximize audience engagement and optimize content promotion. In Morocco, audience measurement has achieved significant advancements, spearheaded by organizations such as the interprofessional media audience center (CIAUMED) and Marocmetrie. Using techniques like watermarking and fingerprinting, Marocmetrie monitors TV consumption for over 1,000 households, providing valuable insights into audience behavior. Despite these advancements, challenges persist-particularly for Laayoune TV, where manual scheduling of daily TV guides often results in misjudged demographic data, inaccurate TV ratings, and suboptimal advertisement pricing. These inefficiencies can lead to financial losses, decreased advertising effectiveness, and diminished viewership, emphasizing the necessity of an automated, data-driven approach for predicting TV program success.