Contextual Bandit Learning for Activity-Aware Things-of-Interest Recommendation in an Assisted Living Environment

Altulayan, May S.; Huang, Chaoran; Yao, Lina; Wang, Xianzhi; Kanhere, Salil S.

doi:10.1007/978-3-030-69377-0_4

Cited by 7 publications

(4 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Healthcare is also an important application area for medical recommendations [ 20 ]. A new algorithm, fb-kNN, is proposed as a recommendation algorithm based on human-disease-rule analysis and then implemented in Healthcare 4.0 for the recommendation of diagnoses and treatments [ 21 ]. A reminder-care system was proposed to help Alzheimer’s patients live safely and independently at home.…”

Section: Related Workmentioning

confidence: 99%

Interpretable Machine Learning for Personalized Medical Recommendations: A LIME-Based Approach

Zhang

Bhatti

et al. 2023

Diagnostics

View full text Add to dashboard Cite

Chronic diseases are increasingly major threats to older persons, seriously affecting their physical health and well-being. Hospitals have accumulated a wealth of health-related data, including patients’ test reports, treatment histories, and diagnostic records, to better understand patients’ health, safety, and disease progression. Extracting relevant information from this data enables physicians to provide personalized patient-treatment recommendations. While collaborative filtering techniques and classical algorithms such as naive Bayes, logistic regression, and decision trees have had notable success in health-recommendation systems, most current systems primarily inform users of their likely preferences without providing explanations. This paper proposes an approach of deep learning with a local interpretable model–agnostic explanations (LIME)-based interpretable recommendation system to solve this problem. Specifically, we apply the proposed approach to two chronic diseases common in older adults: heart disease and diabetes. After data preprocessing, we use six deep-learning algorithms to form interpretations. In the heart-disease data set, the actual model recommendation of multi-layer perceptron and gradient-boosting algorithm differs from the local model’s recommendation of LIME, which can be used as its approximate prediction. From the feature importance of these two algorithms, it can be seen that the CholCheck, GenHith, and HighBP features are the most important for predicting heart disease. In the diabetes data set, the actual model predictions of the multi-layer perceptron and logistic-regression algorithm were little different from the local model’s prediction of LIME, which can be used as its approximate recommendation. Moreover, from the feature importance of the two algorithms, it can be seen that the three features of glucose, BMI, and age were the most important for predicting heart disease. Next, LIME is used to determine the importance of each feature that affected the results of the calculated model. Subsequently, we present the contribution coefficients of these features to the final recommendation. By analyzing the impact of different patient characteristics on the recommendations, our proposed system elucidates the underlying reasons behind these recommendations and enhances patient trust. This approach has important implications for medical recommendation systems and encourages informed decision-making in healthcare.

show abstract

Section: Related Workmentioning

confidence: 99%

Interpretable Machine Learning for Personalized Medical Recommendations: A LIME-Based Approach

Zhang

Bhatti

et al. 2023

Diagnostics

View full text Add to dashboard Cite

show abstract

“…While our focus in this paper is the verification of other parts, we adapt the state-of-the-art approach in this module. Specifically, we propose to utilize recent work propose by Altulyan et al [1], where the approach can effectively suggest the next items to be used by the subject and learns in ambient by both the history of interactions, as well as new interactions, especially when wrong suggestions are made. Despite the proposed method are originally used as part of the reminder system for Alzheimer patients caring, its core function is essentially activity prediction and promotion.…”

Section: Activity Predictionmentioning

confidence: 99%

“…However, inter-device communication and interactions are critical toward building a generalized connected environment, such as a smart home application. Web of Things (WoT) 1,2 is proposed by W3C to facilitate IoT applications, with standardized descriptions of actions, events, and properties of things, as an extension to existing and widely used web protocols. Enabled controls of devices with decoupled specialized APIs, broader interactions between smart things in hypermedia environments can be established comfortably.…”

Section: Introductionmentioning

confidence: 99%

Intent-Aware Interactive Internet of Things for Enhanced Collaborative Ambient Intelligence

Huang

Yao

Wang

et al. 2022

IEEE Internet Comput.

Self Cite

View full text Add to dashboard Cite

Internet of Things (IoT) enables the connection of a broad range of artifacts with advanced sensory technologies and produces massive amounts of data to support ambient intelligence. While the potential of IoT systems is widely recognized, there is still limited work to demonstrate such a system with the autonomy and the ability to execute in the real world. Inspired by the successful introduction of robots to specialized IoT environments, we propose an end-toend solution for a generic, interactive ambient intelligence system, where robotic assistants can assist humans in conducting activities in IoT-enabled smart homes. We evaluate the solution based on implementations of public benchmarks on open-source platforms. We use several activities to demonstrate the effectiveness of the proposed solution in real life.

show abstract

“…In one of our previous works [12], we implemented a prototype system with great consideration of the dynamicity of human activities, which was capable of detecting complex activities. Then, in [13], we presented a Reminder Care System (RCS), which, in addition to being able to learn the dynamicity of human activities, could also remind patients about their needs correctly, without requiring their feedback. The problem was formulated using a contextual bandit approach, which considers contexts as input to recommend the next action.…”

Section: Introductionmentioning

confidence: 99%

Context-Induced Activity Monitoring for On-Demand Things-of-Interest Recommendation in an Ambient Intelligent Environment

et al. 2021

Self Cite

View full text Add to dashboard Cite

Recommendation systems are crucial in the provision of services to the elderly with Alzheimer’s disease in IoT-based smart home environments. In this work, a Reminder Care System (RCS) is presented to help Alzheimer patients live in and operate their homes safely and independently. A contextual bandit approach is utilized in the formulation of the proposed recommendation system to tackle dynamicity in human activities and to construct accurate recommendations that meet user needs without their feedback. The system was evaluated based on three public datasets using a cumulative reward as a metric. Our experimental results demonstrate the feasibility and effectiveness of the proposed Reminder Care System for real-world IoT-based smart home applications.

show abstract

Contextual Bandit Learning for Activity-Aware Things-of-Interest Recommendation in an Assisted Living Environment

Cited by 7 publications

References 26 publications

Interpretable Machine Learning for Personalized Medical Recommendations: A LIME-Based Approach

Interpretable Machine Learning for Personalized Medical Recommendations: A LIME-Based Approach

Intent-Aware Interactive Internet of Things for Enhanced Collaborative Ambient Intelligence

Context-Induced Activity Monitoring for On-Demand Things-of-Interest Recommendation in an Ambient Intelligent Environment

Contact Info

Product

Resources

About