YEREVAN, Armenia — The American University of Armenia (AUA) Zaven P. and Sonia Akian College of Science and Engineering is proud to announce that a team of AUA alumni and faculty has coauthored a research paper published in the Journal of Universal Computer Science (JUCS).
The article, titled “Interpretable Clustering Using Dempster-Shafer Theory,” is the result of collaboration among several AUA scholars: recent Bachelor of Science in Data Science graduate and current Master of Science in Computer and Information Science student Hovhannes Hovhannisyan (BSDS ’25, MSCIS ’27); fellow alumnus Aram Adamyan (BSDS ’24); Dr. Nelson Baloian, visiting associate professor at AUA and faculty member at the University of Chile in Santiago; and Ashot Harutyunyan, former AUA adjunct lecturer and now head of the Artificial Intelligence Lab at Yerevan State University. They worked alongside Daniel Redrigan of the University of Chile.
JUCS is a long-standing, Scopus- and Web of Science–indexed journal, also listed in DOAJ and UGC CARE. With nearly three decades of continuous coverage (1996–2025), an H-index of 59, and strong recognition in computer science, the journal provides a rigorous and credible platform for impactful research.
The article introduces DSClustering, a novel algorithm that combines clustering accuracy with interpretability. Unlike traditional methods, such as K-means or DBSCAN, which often lack transparency, DSClustering applies belief functions from Dempster-Shafer theory to provide rule-based explanations for cluster assignments. Tested on real-world datasets, the algorithm proved both competitive in performance and more transparent in its outputs, making it especially for fields such as healthcare, finance, and consumer analytics where clear decision-making is essential.
“Beyond academia, our work on DSClustering addresses real-world needs in areas such as customer segmentation, decision explanation, and uncertainty quantification — applications relevant to healthcare, finance, retail, and public policy,” Hovhannisyan emphasized. “In an era where much focus is on deep learning and generative AI, this research highlights the importance of interpretability and trust in AI systems, helping to minimize uncertainty in decision-making.” He also invited anyone interested in making their operations more measurable and turning explainable AI into tangible revenue growth to connect with the team for assistance.
The publication underscores both the innovative contributions of AUA’s graduates and faculty to the field of AI and the University’s growing role in international, cross-institutional research collaborations.
Founded in 1991, the American University of Armenia (AUA) is a private, independent university located in Yerevan, Armenia, affiliated with the University of California, and accredited by the WASC Senior College and University Commission in the United States. AUA provides local and international students with Western-style education through top-quality undergraduate and graduate degree and certificate programs, promotes research and innovation, encourages civic engagement and community service, and fosters democratic values.