Computational Communication Intelligence: Exploring Linguistic Manifestation and Social Dynamics in Online Communication

02 Dec
Monday, 12/02/2013 4:30am to 6:30am
Ph.D. Thesis Defense

Xiaoxi Xu

Computer Science Building, Room 151

We live in an age of online communication. As social media becomes an integral part of our life, online communication becomes an essential skill. In this research, we aim to understand how people can communicate effectively online. We research components of success in online communication and present scientific methods to study the skill of effective communication. This research contributes toward the fields of machine learning and communication studies.

We pioneered the study of a communication phenomenon we call communication intelligence in online interactions. We created a theory about communication intelligence that includes measuring participants' ten high-order communication skills, including restraint, self-reflection, perspective taking, and balance. We presented a multi-perspective analysis for understanding communication intelligence, including its diverse language, universal linguistic characteristics, social dynamics, and the effects of communication modality on communication intelligence.

In the area of machine learning, we contributed new computational models and formulations for addressing multi-label and multi-task machine learning problems. We developed a new hierarchical probabilistic model for addressing the problem of simultaneously identifying multiple intelligence-embodied communication skills from natural language. The model learns the topic assignment for each sentence and provides a practical and simple way to determine document labels without relying on a threshold function. The model performance increases as the number of labels grows, which makes it a promising approach for large-scale data analysis. We also developed a new multi-task formulation for simultaneously identifying multiple intelligence-embodied communication skills using lexical, discourse, and interaction features. The key merit of this model is that it is a general multi-task formulation that unifies many well-known regularization techniques, including Lasso, group Lasso, sparse-group Lasso, and the Dirty model. Moreover, it can be applied to streaming data to perform large-scale analysis in real time.

Advisor: Beverly Woolf