AI & Service Science

by Robin Qiu

---

 

 

 

 

 

Additional details and published articles on Dr. Qiu’s research topics can be found here.

 

1. Health and Healthcare Analytics

   • Directed Disease Networks to Facilitate Multiple-Disease Risk Assessment Modeling— This research investigates multiple disease risk prediction modeling, aimed at assessing future disease risks for an individual who is ready for discharge after hospitalization, proposing a novel framework that combines directed disease network and recommendation system techniques to substantially enhance multiple disease risk predictive modeling. The proposed framework can also be used to develop innovative tools that helps individuals create and maintain better, long-term healthcare plans.
   • Patient-centered Deep Learning Model and Diagnosis Service for Persons with Alzheimer’s Disease— Because pharmaceutical companies have yet to develop cures and treatments for Alzheimer’s Disease (AD), early detection and intervention is the best choice of improving future patients’ quality of life. Thus, developing patient-centric predictive models and enabling self-diagnosis services are of great potential. This research examines the fast development of big data technologies and machine learning methods, including early detection tools that could make a difference in discovering non-pharmacologic therapy solutions to slow AD progression.

2. Smart Cities;

   • Leveraging Big Data Platform Technologies and Analytics to Enhance Smart City Mobility Services— The Internet of Things (IoT) allows objects to be sensed and managed over networks, creating opportunities for beneficial interaction and integration between the physical world, computer-based systems, and human beings. The recently enabled people-centric sensing or social sensing transforms how we sense and interact with the world. For instance, social sensing via mobile apps complements physical sensing (e.g., IoT) by substantially extending the horizon we know about our living communities and environments in real time.
   • Developing a smart service system to enrich bike riders’ experience— Social sensing via mobile apps complements physical sensing (e.g., IoT) by substantially extending the horizon we know about our daily life in real time. This research focuses on integrating physical and social sensing to enable better and smarter services and collecting, processing, and modeling data on riders’ profiles and mobility needs.
   • How to Rebalance Shared Electric Vehicles? An Integrated Price-incentive Policy— Recently, electric vehicle sharing (EVS) has received increasing attention around the world to ease traffic congestion and environmental. This research investigates an integrated strategy by combining the price-incentive approach with the trip-selection approach and model uncertain travel demands in a two-stage process based on the integrated strategy.

3. Education Services and Systems;

   • A quantitative and model-driven approach to assessing higher education in the United States of America— University ranking or higher education assessment in general has been attracting more and more public attention over the years. However, the subjectivity-based evaluation index and indicator selections and weights that are widely adopted in most existing ranking systems have been called into question. In other words, the objectivity and impartiality of those rankings has been worrisome. To address these concerns, this research investigates a quantitative and model-driven approach to acquiring the evaluation index and indicator weights in the US News & World Report ranking system.
   • A Big Data-based Smart Evaluation System using Public Opinion Aggregation— Assessing service quality proves very subjective, varying with objectives, methods, tools, and areas of assessment in the service sector. Customers’ perception of services usually plays an essential role in assessing the quality of services, so mining customers’ opinions in real time may help capture and decipher perception of service experiences. This research reviews big data-based framework in support of data retrieving, aggregations, transformations, and visualizations by focusing on public ratings and comments from different data sources.

4. Big Data and Analytics:

   • Noninvasive MapReduce Performance Tuning Using Multiple Tuning Methods on Hadoop— There are more than 190 configuration parameters affecting the performance of MapReduce jobs on Hadoop. A self-tuning system to improve MapReduce’s performance in an automated and efficient manner in a complicated Hadoop environment is needed. This research explores multiple tuning methods to improve tuning efficiency for MapReduce performance on Hadoop. The proposed Catla system employs succinct templates and proper schemes of MapReduce algorithms, which can be incorporated in facilitating the tuning and optimization of MapReduce performance.