What is it?
The Rol-AI-Dex will be an application that showcases the depth, breadth, and growing connectivity of AI-relevant work across the Pitt community. Through an interactive network visualization, users can see how people, projects, organizations, and products are connected, revealing the rich structure of Pitt’s AI ecosystem. The application will also support searching and filtering by different types of entries, allowing the visualization to update dynamically and highlight areas of interest.
Who will use it?
The Rol-AI-Dex is a centralized, living inventory of people, projects, datasets, and initiatives related to AI across the University. It enables Pitt to see where AI-relevant work is happening, identify gaps and overlaps, support strategic planning, and uncover connections across disciplines. For faculty, staff, and students, the Rol-AI-Dex makes it easy to find AI expertise, discover potential collaborators, and share resources and knowledge across the community.
What is our process?
- Step 1: Ontology - Define What We Track and How It Connects
-
We start by agreeing on the basic building blocks of the Rol-AI-Dex: the kinds of things we track (people, organizations, projects, and products) and the ways they can be connected (for example, who leads a project, who collaborates on it, or which organization someone belongs to).
This shared structure ensures all information is entered consistently and allows the Rol-AI-Dex to accurately map, search, and reveal connections across AI activity at Pitt.
- Step 2: Data Collection and Organization
-
We started with AI-related information that was collected last semester and stored in a single Google Sheet. Because the sheet wasn’t originally designed for the Rol-AI-Dex structure, we reorganized that information into the formats needed for people, organizations, projects, products, and their relationships.
From this, we created separate files for “things” (nodes) and “connections” (edges). The system then reads these files and builds the network graph that powers the Rol-AI-Dex.
- Step 3: Visualize the Network
- We use a visualization tool that works well with our network graphs to turn the Rol-AI-Dex data into interactive visuals. This allows people to explore AI activity as a network of connected people, projects, and organizations rather than a static list.
- Step 4: Search Functionality
-
We are testing and will build a search feature that allows users to choose a starting point (for example, a person or project) and specify how far out into the network they want to explore. The system will then find and display nearby connections as an interactive network visualization. Initially, this functionality will be available internally as we continue to develop and refine the experience.
- Step 5: Build the External Application
-
We will create a public-facing application that allows people to explore the Rol-AI-Dex through an interactive interface. This application will include features like search boxes, drop-down menus, and filters to make it easy to find and explore information.
Over time, the text search box will connect directly to our internal search functionality so visualizations update automatically based on what a user enters.
Where are we now, and what are the next steps?
We have built an early version of the Rol-AI-Dex using our existing campus data to create a first map of people, projects, organizations, and products related to AI at Pitt. This version helps us see what’s already visible—and where important work may be missing.
Now we need your help to make it better.
-
If you see yourself, your project, or your organization:
Please click the 👍 “I’m represented” button so we know the information looks correct. -
If you do NOT see yourself or your work:
Please fill out our short input form to add your information.
Your feedback directly shapes the Rol-AI-Dex. The more people who confirm or add their information, the more accurate, useful, and representative this campus-wide map becomes.