Prasan Priyadarshana Yapa Wins Best Paper Award at WEBIST 2023

Prasan Priyadarshana Yapa, a second-year doctoral student at the KUAS Graduate School of Engineering and member of the Ubiquitous and Personal Computing Laboratory (specializing in ubiquitous computing, personal informatics, and digital health, directed by Dr. Zilu Liang), has received the Best Paper Award from the 19th International Conference on Web Information Systems and Technologies (WEBIST 2023) held in Rome, Italy, for his paper titled “Who Says What (WSW): A Novel Model for Utterance-Aware Speaker Identification in Text-Based Multi-Party Conversations” (co-authored by Dr. Zilu Liang and Dr. Ian Piumarta). This award recognizes researchers both for their “outstanding research” and for their “best oral presentation” delivered at the conference. The WEBIST conference had an acceptance rate of as low as 25% for full papers this year, and the Best Paper Award honors only the top 1% of exceptional papers.

The award-winning paper proposed a novel computational method for speaker identification in text-based multi-party conversations. The method adapts pre-trained deep-learning language models for utterance structure modeling and utterance semantic modeling based on the relative semantic distance of utterances. The method was validated on four large benchmark datasets and achieved a new state-of-the-art performance. This work will pave the way for the development of new depression screening methods based on multi-party SNS chat log analysis. 

(Dr. Zilu Liang, Faculty of Engineering)