Date: April 9, 2018
Time: 9:35 am
Location: Room 252, CSE Building
Admission: This event is free and open to the public.
Style-Aware Representations of Text
Abstract: Style in written language refers to the way authors choose to structure text. It includes verbosity, which can be measured in terms of sentence length and/ or paragraph length. Style also includes elements of syntax, for example to account for the use of passive versus active voice.
Other relevant elements of style are word choice and rhythm. During this talk I will present the work I have done related to capturing style in text, from typed-character n-grams, to emotion flow in text, to the more recent representation learning approaches. I will show results from using these different ideas in a task where style is key to achieve state of the art performance: prediction of books likeability.
To conclude I will discuss our ongoing work in this project.
Biography: Thamar Solorio is an associate professor in the Department of Computer Science at the University of Houston (UH). She is the founder and director of the Research in Text Understanding and Analysis of Language (RiTUAL) group at UH. Her main research interests include stylistic modeling of text, syntactic analysis of mixed language data, language assessment, and information extraction from social media data. She has M.S. and Ph.D. degrees in Computer Science from INAOE, Puebla, Mexico. The Department of Defense and the National Science Foundation currently funds her research program. She is the recipient of a CAREER award for her work in authorship analysis, and the 2014 Denice Denton Emerging Leaders ABIE Award. She serves as editorial board member for the Journal on Artificial Intelligence Research (JAIR) and the Computer Speech and Language Journal.