Events
Sessions promoted by the IDeaS Group in external events.
PDW at the Academy of Management 2020 Conference
Title: Topic Modeling Advances - Curating Corpora, Using Structured Models, and Theorizing with Visuals
Speakers and Facilitators: Hannigan, Haans, Tchalian, Valadao, Glaser & Jennings
Description: This session built on our 2017 & 2018 topic modeling PDWs and was offered in "real-time" mode at the virtual 2020 AOM Conference. The workshop focused on building theory with visual artifacts created from topic modeling methods. Practices of visualization are important in both the doing and understanding of big data analytics. In our view of an interpretive approach to rendering data, models and theory in social science, visual artifacts are boundary objects that are central in the process. This PDW was an interactive session, with brief method- and theory-building statements, followed by specific exercises in breakout sessions with participants. The groups actively engaged in generating novel theorizations based on pre-rendered visualizations.
PDW at the Academy of Management 2018 Conference
Title: Topic Modeling - A Hands-on Approach to Analysis and Theory Building
Speakers and Facilitators: Hannigan, Haans, Tchalian, Vakili, Glaser, Wang, Kaplan & Jennings
Description: Whereas the 2017 PDW was focused primarily on the technique and details of topic modeling, this PDW focused on types of topic modeling, specific applications, and building theory from them. More importantly, this PDW discussed topic modeling as a non-linear process, in which the researcher iterates between three specific stages: (1) Rendering Corpora (selecting and trimming), (2) Rendering Topics (applying algorithms and fitting), and (3) Rendering Theoretical Artifacts (creating and building theory).
PDW at the Academy of Management 2017 Conference
Title: Topic Modeling in Quant and Qual Research: A Hands-On Approach
Speakers and Facilitators: Jennings, Kaplan, Hannigan, Haans, Wang, Tchalian & Vakili
Description: This PDW engaged participants in the theory and practice of topic modeling using a hands-on approach. Participants where guided to run pre-built programs based on standard unstructured topic modeling routines in R — which were applied to various textual datasets — and then delved into the principles behind such analyses. Additionally, participants had the chance to learn and discuss pathways for theorization and write-up targeting publication of articles in management top journals.