The School of Mathematical and Computational Sciences welcomes the campus community to public research and teaching presentations by Dr. Paul Sheridan, candidate for the tenure track position in Data Science. All are welcome to attend.

Research Talk Title:  "The Literary Theme Ontology with Applications to Media Annotation and
Information Retrieval" 10am-11am

Abstract:  Literary themes (e.g. courage, capitalism, and coming of age) make for a natural organizing hub when it comes to the study of cultural evolution. Inspired by the Gene Ontology knowledgebase resource from bioinformatics, the Literary Theme Ontology (LTO) is the first ever controlled vocabulary of literary themes that are hierarchically arranged into a directed acyclic graph structure. The LTO forms the core knowledgebase for an open access, community-based fiction studies project that classifies literary themes through the use of a hierarchically structured controlled-vocabulary, and annotates works of fiction with the themes within a collaborative framework. I will talk about the current state of this emerging line of research of mine with particular emphasis on the ecosystem of downstream statistical analysis we are in the process of developing. Finally, I will outline my vision of capitalizing on converging developments in interoperable frameworks of controlled vocabularies, online collaborative platform infrastructures, and modern data science techniques to build up a large-scale database of
thematically annotated works of fiction that researchers will be able to use to test their various hypotheses about cultural evolution.

Teaching Talk titled:  "Model Selection in Bayesian Multiple Linear Regression" 12:30pm-1:30pm

Abstract:  I will talk about model selection in a Bayesian approach to multiple linear regression using the Bayesian information criterion, or BIC. While my talk is directed toward upper level undergraduate students, anyone who is acquainted with Bayes' theorem and has performed a multiple linear regression
or two will be able to get the gist of it. And even if terms like "Bayesian", "information criterion", and "multiple linear regression" leave you scratching your head, rest assured that you'll leave my talk with a conceptual understanding of what "model selection" is all about. In other words: drop on by, there's something in it for anyone at the university level who is interested in probability and statistics.

Thursday, May 6- Research Talk 10am-11am/Teaching Talk 12:30pm-1:30pm

Via Zoom
https://zoom.us/j/98875873796?pwd=RC9yaTlWdEU3aWtKTHluU1ovTGc5UT09

Meeting ID: 988 7587 3796
Passcode: 218422