Text as Data Practice Group

The Text as Data Practice Group offers monthly gatherings focusing on specific computational tools or methods that we will learn and work through together. The group is discipline-agnostic, and all levels are welcome to experiment and learn from the tutorials at hand, and from each other. The group operates in the spirit of creating community, nurturing peer-to-peer learning, and exploring emerging text and data analysis methods on campus.  

Open to students, staff, faculty, and all curious parties. Bring a laptop!

Facilitated by: Cody Hennesy, Alexis Logsdon, and Wanda Marsolek.

We will reconvene in Fall 2020 - sign-up for our mailing list here!

Summer 2020

Below are some options available for self-paced online learning in areas related to text mining over the Summer!

Digital Scholar Workbench (JSTOR)

Online lessons in scholarly text analysis methods (e.g. word frequencies, TF/IDF) using JSTOR data, as well as beginner's lessons for Jupyter notebooks and Python.

Coursera (thru July 31, 2020)

Instructions for connecting to Coursera via UMN

LinkedIn Learning

Login to LinkedIn Learning to start 

Previous meetings

Spring 2020 Meetings

Text as Data Lightning Talks (Round Two)
Wed, May 13, 3-4pm  (Online)

The final meeting of our inaugural semester will feature another round of lightning talks from UMN researchers using "text as data" methods in their own scholarship. 

  • Mariya Gyendina (UMN TC, Libraries) - Consulting based on who I think you are: Mixed methods study of feedback in an online writing center
  • Siyu Li (UMN TC, Political Science) - Evolving Norm of Collegiality: Analyzing the Sentiments of Supreme Court Oral Arguments
  • Ted Pedersen (UMN Duluth, Computer Science) - Automatic Detection of Hate Speech and Islamophobia

Text as data lightning talks
Wed, April 15, 3-4pm (Online)

Join us for a series of short lightning talks from staff and graduate students about their own text as data research. From issues in data sourcing and cleaning, to working with methods and interpreting results, this will be an opportunity to discuss possible approaches and solutions with peers across the disciplines.

  • Kelsey Neis (Office of Info Technology) - Text as data as text adventure
  • Neeraj Rajasekar (Sociology) - Diversity discourse in the news: A Quantitative content analysis of "diversity" in political news media
  • Cody Hennesy & David Naughton (Libraries) - Computational analysis of Library Quarterly

Sentiment Analysis for Exploratory Data Analysis
Wed, Feb 19, 2020, 3-4pm (Wilson Library Collaboration Studio)
Join us for our first meeting, where we’ll work through the Programming Historian’s tutorial on Sentiment Analysis in Python. Sentiment Analysis is a form of natural language processing that seeks to quantify the emotional intensity of words and phrases within a text or texts. This tool can be helpful for analyzing interview transcripts, newspaper articles on a specific topic, or even poetry! 

Installation and laptop set-up: if you can already use Python on your laptop, you should be all set. If you’re new to Python, we recommend installing Anaconda for Python 3.7. If you need any installation help, you can email Cody (chennesy@umn.edu) or stop by on Feb 19 at 2:30 for hands-on help!

Last Updated: May 20, 2020 10:00 AM