简体中文
EnglishWhat is GroupLens at the University of Minnesota?
Updated
GroupLens is a research center at the University of Minnesota focused on social and human-centered computing. It advances the theory and practice of social computing by building and studying systems that real people actually use, and it publishes research primarily in computer science but also in psychology, sociology, and medicine. If you want to understand the lab behind MovieLens, find its publications or datasets, or explore its open-source recommendation tools, this is the organization to know.
What GroupLens does
The center describes its mission as advancing social computing “by building and understanding systems used by real people.” That combination matters: GroupLens is not only a theory group. It builds working systems, runs experiments on them, and reports what happens.
Its research output appears in conferences and journals, mainly in computer science, with contributions extending into psychology, sociology, and medicine. The site points readers to a blog for research highlights and a publications page for a comprehensive view of contributions.
Featured research areas
Recent work highlighted on the site shows the range of topics:
- Crowd-powered therapy. A blog post on the Flip*Doubt prototype, an Honorable Mention paper at CSCW 2021, asks how technology can expand delivery of clinically validated therapeutic techniques when there are not enough trained mental health professionals. This is a research direction, not a clinical service.
- Bots on Wikipedia. A post from the University of Minnesota’s “human-centered computing for social good” REU examines conflict among bots that edit Wikipedia.
These examples illustrate the lab’s pattern: pick a real social system, study how people and automation interact within it, and publish the findings.
Featured projects
GroupLens has built and studied real systems since the release of MovieLens in 1997. Three featured projects show what that looks like in practice:
| Project | What it is | What it demonstrates |
|---|---|---|
| MovieLens | A website that helps people find movies to watch, with hundreds of thousands of registered users | Online field experiments in automated content recommendation, recommendation interfaces, tagging-based recommenders, member-maintained databases, and intelligent user interface design |
| Cyclopath | A geowiki: an editable map where anyone can share notes about roads and trails, add tags for special locations, and fix map problems like missing trails | Community-maintained geographic data; described as the most comprehensive and up-to-date bicycle information resource in the world, used by hundreds of Twin Cities cyclists |
| LensKit | An open-source toolkit for building, researching, and studying recommender systems | Practical recommender infrastructure for projects and research |
MovieLens is the clearest example of the lab’s method: because it operates a real site with real users, GroupLens can run field experiments rather than only lab studies. Cyclopath applies the same community-contribution idea to geographic data. LensKit turns recommender research into reusable tooling.
Who the site is for
The site serves several audiences:
- Researchers who want publications, datasets, or collaboration context in social computing and recommender systems.
- Developers looking for an open-source recommender toolkit (LensKit).
- Students and prospective collaborators exploring the lab’s projects and REU work.
- Curious users who want to understand the organization behind MovieLens.
How to use the site
Start from the navigation: about, datasets, publications, and blog. Use the publications page for a comprehensive view of research contributions, the blog for accessible highlights of recent papers, and the projects page for a full list of active projects. If your interest is recommender systems specifically, LensKit is the most directly usable output; if your interest is research findings, the publications and blog are the entry points.
One caveat: the site presents research and project descriptions. It does not present itself as offering consumer services beyond the projects it runs, and details like pricing or access terms for any tool are not stated on the pages summarized here, so check the individual project pages before relying on them.