Text Analysis

The Text Analysis in a World of AI (TAWAI) working group brings together scholars, students, and staff from across campus who share in interest in computational text analysis. Participants range from expert practitioners and experienced teachers to novices just getting started.

We are focused on performing text analysis in a World of AI. This community of practice meets once a month to:

• discuss, examine, and collaboratively experiment with new tools, models, methods, and other developments in this field. Tools range from text databases to out-of-the-box software, text analysis programming packages, and both open-source and proprietary Generative AI tools.
• share participants' own progress and work.

Such meetings are designed to be inter- and trans-disciplinary by engaging materials and activities that could be adapted and used by researchers in diverse fields across the sciences, arts, humanities, and vocational fields (like engineering and business).

While welcoming to all relevant approaches, this group will focus on applying and adapting text analysis methods and tools (integrated with language models) to answer domain-specific research questions (rather than developing the tools or models themselves).

We meet once a month (usually the last Thursday). See the "Join / Communicate" section below for more information how to join and get more information.

In a rapidly changing technology / knowledge / information / communication landscape, a few things are clear about the world we live in:

1. We are awash with data.
2. Much of this data is in text form (aka. “natural language” data): websites, blogs, reports, social media posts, academic articles, news articles and transcripts, song lyrics, etc.
3. In the twenty-first century, computational text analysis and Natural Language Processing (NLP) techniques have provided new avenues for studying this textual data
4. The recent rise of Generative AI tools is transforming how text can be produced and analyzed. It is thus also transforming how we read, analyze, and write texts and how we conduct research, share information, and judge the accuracy and validity of this information.
5. We live in a world where the rise of cheap-to-produce computer-generated texts is blending with a social and traditional media landscape already flooded with dis- and misinformation. Thus, we must better understand this proliferation of misinformation and develop the tools to combat it.

This community of practice also aligns well with some priorities and initiatives at Dartmouth College:

1. Dartmouth College’s efforts to promote AI research and teaching
2. The Dartmouth Libraries’ strategic pillars or goals to:
   1. Accelerate advanced research
   2. Elevate scholarship with powerful research tools and methods

The principal communication platform for this group is through an email list and our Slack workspace. Through these platforms, you can learn about upcoming events, find and share resources, ask and answer questions.

To join the email list: email Jeremy Mikecz (email address available in Dartmouth outlook) and ask to be added to the list.

To join the Slack workspace:

1. Open the Slack app (if you're not already on Dartmouth's Slack, follow the directions here).

2. select the “+” (Add Workspaces) option on the lefthand toolbar

3. Select “Find more Dartmouth College workspaces” (if you use the standalone Slack app you may be prompted to sign on to Slack online to continue)

4. Search for “text” and choose “Text Analysis Working Group”.

5. Select "View Details" and "Request to Join".

In our monthly meetings we engage in a variety of collaborative and hands-on activities. All group members are invited to proposed or facilitate new sessions. Some possible activities include:

1. Demo & Practice Workshops: A member of the community (or a visitor) presents a new tool, method, model, or approach to Text Analysis (with or without Gen AI / Large Language Models)
2. Idea Brainstorming and Feedback Sessions: Participants solicit ideas and suggestions for how to resolve a particular problem related to text analysis, language models, and Generative AI. Or participants come to discuss a brand new tool or method, examine and experiment with it, and brainstorm possible use cases of that tool.
3. Collaborative projects: Learn new skills while participating in a collaborative project (that, in some cases, may produce a real product for the community)
4. Hackathons: the group works together to create a tool or script that accomplishes a particular assigned task or solves a specific problem.
5. Review Sessions: members of the group can share works in progress (articles, code, presentations, job talks) related to the themes of the group and solicit feedback from a friendly and supportive audience.
6. Mob Learning: no leader; start with some code / coding notebook, tutorial, or recorded lecture and work our way through it together.