Spoiled for Choice? Personalized Recommendation for Healthcare Decisions: A Multiarmed Bandit Approach

Abstract

Choice overload is a common problem in many online settings, including healthcare. Online healthcare platforms tend to provide a large variety of behavior intervention information or programs to help individuals modify their lifestyles to improve wellness. However, having too many options can significantly increase searching cost, prevent users from discovering the truly relevant interventions, and harm users’ long-term healthcare decision-making efficiency. This motivates us to propose a personalized healthcare recommendation system to provide tailored support for individuals’ intervention participation. The proposed framework, a deep-learning and diversity-enhanced multiarmed bandit (DLDE-MAB), integrates several predictive and prescriptive analytics components to combat the unique challenges presented in the healthcare recommendation setting. It leverages online machine learning to provide adaptive and real-time support, a theory-guided diversity promotion scheme to cover multiple healthcare needs, and deep learning to further enhance dynamic context representation. Through extensive experiments, we show that the proposed framework outperforms various competing models in terms of its adaptivity to data dynamics, diversity, and uncertainty. The proposed model and evaluation results provide important implications for business intelligence and personalized, contextualized, and agile healthcare decision making.

A Analytical and Practical Insights into Multi-Armed Bandit Problems in Recommendation Systems

“…Besides, because of the balance of MAB algorithm between exploitation and exploration, it can deal with the cold-start problem in recommendation systems effectively [12]. What is more, MAB algorithm does not only consider global optimization but also attaches importance individual feedback thus can offer more personalized recommendation service [13].…”

Personalized Dynamic Pricing Based on Improved Thompson Sampling

“…The MAB problem refers to the challenge of selecting optimal actions to maximize cumulative rewards within limited periods, with the core issue being the balance between exploration and exploitation. Recently, the MAB framework has gained widespread application in various fields, such as recommendation systems [36,37], healthcare [38], and dynamic pricing [39][40][41][42]. Currently, three commonly used algorithms for solving the MAB problem are the ε-greedy algorithm, Upper Confidence Bound (UCB) algorithm, and TS.…”

Advancing decision-making strategies through a comprehensive study of Multi-Armed Bandit algorithms and applications

“…In practical applications, MAB algorithms enhance personalization in e-commerce, streaming services, and news delivery by adapting recommendations in real time to user preferences [5]. In healthcare, they are utilized for optimizing clinical trials, drug dosages, and treatment plans, balancing well-known treatments with experimental options [6]. In finance, MAB strategies support portfolio management, dynamic pricing, and advertising bidding to optimize returns while managing risks [7].…”