College of Science & Engineering
This group's research spans the fields of high-performance computing, recommender systems, learning analytics, text analytics, and graph neural networks.
- Research in the area of high-performance computing focuses on the design and implementation of algorithms for analyzing large-scale datasets that arise in machine learning tasks and graph analytics. These tools are used extensively in fields such as anomaly detection, precision healthcare, and recommender systems.
- Research in the area of recommender systems focuses on the design and development of algorithms to improve the quality of recommendations served to a user of the system. After thoroughly exploring several large-scale datasets, these researchers have identified certain fundamental characteristics that affect the performance of existing recommendation schemes. This has led to the development of new methods for Top-N recommendation methods that outperform the state-of-the-art while also being efficient and readily applicable in large-scale settings. The group intends to push Top-N recommendation performance even further by considering new and efficient personalized models as well as to extend their methods to take into account meta-information of the items. Further, they will explore the utility of the user-item interaction graphs to further improve recommendation quality.
- Research in the area of learning analytics focuses on the development of predictive models for estimating the performance (i.e., grades) of students on future courses, ranking models for Top-N course recommendation, identifying enrollment patterns that are associated with course success or failure, and studying curriculum planning in terms of course timing and ordering and how it relates to the student’s academic performance and graduation time. These models aim to help students make informed decisions about which courses to register for and help them with course sequencing, which can improve student retention and lead to successful and timely graduation.
Research in the area of text analytics focuses on the design and development of distant supervised algorithms, with applications to text-segmentation, information retrieval, and citation analysis. The algorithms do not depend on explicitly labeled data for training, but leverage other sources (such as bag of labels associated with a group of data points) as a source of distant supervision. The algorithms aim to decrease the reliance on the labeled training data in the applications where generating such labels is expensive, time-consuming and the meaning of the labels is user-defined.
Research in the area of graph neural networks focuses on the development of deep learning methods that effectively encode the topological structure of graphs into node embeddings and graph embeddings, and application of graph neural networks in computational chemistry and materials simulation. The researchers aim to improve existing methods by combining the expressibility of deep neural networks with topological information within graphs.