<p>Adaptive Attention-Based Marketing: A Review of Emotion- and Context-Aware Approaches for Mobile App Engagement Optimization</p>

With the rise in mobile application usage, competition for user attention has become more intense, and the demand for intelligent and personalized engagement strategies has grown. Despite the significant progress made in emotion recognition, context-aware computing, and adaptive recommendation systems, the research is still scattered in many different fields. This review aims to fill this gap by exploring the new paradigm of Adaptive Attention-Based Marketing (AABM) that combines real-time emotion detection, context-aware sensing, and adaptive decision-making to improve mobile app engagement. The literature review was carried out in a structured manner using the Scopus database, and the period of literature search was from 2018 to 2025. There were 408 records identified and screened, 40 studies were included in the final qualitative synthesis. The review covers advances in multimodal emotion recognition, context recognition, and Artificial Intelligence (AI) for personalization, such as machine learning, deep learning, contextual bandits, and reinforcement learning. The results show that three key enablers for AABM are multimodal sensing, contextual intelligence, and adaptive learning algorithms. Its applications span e-commerce, gamified systems, mobile notifications, and the healthcare sector, showcasing its ability to enhance engagement, retention, and user experience. However, challenges remain regarding privacy, transparency, multimodal data availability, and long-term evaluation. This review brings together existing knowledge and identifies research gaps, as well as directions for the development of privacy-aware and ethically responsible adaptive engagement systems.