GROUPLENSABOUTDATASETSPUBLICATIONSBLOGPROJECTSCONTACT US MovieLens is a web site that helps people find movies to watch. It has hundreds of thousands of registered users. We conduct online field experiments in MovieLens in the areas of automated content recommendation, recommendation interfaces, tagging-based recommenders and interfaces, member-maintained databases, and intelligent user

interface

MOVIELENS | GROUPLENSMOVIELENS 10M DATASETMOVIELENS 100K DATASETMOVIELENS TAG GENOME DATASET MovieLens 1B Synthetic Dataset. MovieLens 1B is a synthetic dataset that is expanded from the 20 million real-world ratings from ML-20M, distributed in support of MLPerf.Note that these data are distributed as .npz files, which you must read using python and numpy.. README DATASETS | GROUPLENS Datasets. GroupLens Research has collected and made available several datasets. Choose the one you’re interested in from the menu on the right. Before using these data sets, please review their README files for the usage licenses and other details. MOVIELENS 100K DATASET MovieLens 100K movie ratings. Stable benchmark dataset. 100,000 ratings from 1000 users on 1700 movies. Released 4/1998. README.txt

ml-100k.zip (size:

MOVIELENS 1M DATASET MovieLens 1M movie ratings. Stable benchmark dataset. 1 million ratings from 6000 users on 4000 movies. Released 2/2003. README.txt ml-1m.zip (size: 6 MB, checksum) Permalink:

HETREC 2011

The 2nd International Workshop on Information Heterogeneity and Fusion in Recommender Systems (HetRec 2011, has released datasets from

Delicious, Last.fm

EVALUATING COLLABORATIVE FILTERING RECOMMENDER SYSTEMS 6 • J. L. Herlocker et al. the most successful technologies for recommender systems, called collabora- tive filtering, has been developed and improved over the past decade to the point where a wide variety of algorithms exist for generating recommenda- BEING ACCURATE IS NOT ENOUGH: HOW ACCURACY METRICS HAVE Being Accurate is Not Enough: How Accuracy Metrics have hurt Recommender Systems Sean M. McNee Abstract Recommender systems have shown great potential to COMBINING COLLABORATIVE FILTERING WITH PERSONAL AGENTS FOR Combining Collaborative Filtering with Personal Agents for Better Recommendations Nathaniel Good, J. Ben Schafer, Joseph A. Konstan, Al

Borchers,

DO YOU TRUST YOUR RECOMMENDATIONS? AN EXPLORATION OF 3 Model Builder (2) Predictor (4) Personal Information (1) Data Store (3) Predicted Preferences (5) Recommender (centralized or distributed) Fig.1. Conceptual model of the interaction between a user and a

recommender system.

GROUPLENSABOUTDATASETSPUBLICATIONSBLOGPROJECTSCONTACT US MovieLens is a web site that helps people find movies to watch. It has hundreds of thousands of registered users. We conduct online field experiments in MovieLens in the areas of automated content recommendation, recommendation interfaces, tagging-based recommenders and interfaces, member-maintained databases, and intelligent user

interface

MOVIELENS | GROUPLENSMOVIELENS 10M DATASETMOVIELENS 100K DATASETMOVIELENS TAG GENOME DATASET MovieLens 1B Synthetic Dataset. MovieLens 1B is a synthetic dataset that is expanded from the 20 million real-world ratings from ML-20M, distributed in support of MLPerf.Note that these data are distributed as .npz files, which you must read using python and numpy.. README DATASETS | GROUPLENS Datasets. GroupLens Research has collected and made available several datasets. Choose the one you’re interested in from the menu on the right. Before using these data sets, please review their README files for the usage licenses and other details. MOVIELENS 100K DATASET MovieLens 100K movie ratings. Stable benchmark dataset. 100,000 ratings from 1000 users on 1700 movies. Released 4/1998. README.txt

ml-100k.zip (size:

MOVIELENS 1M DATASET MovieLens 1M movie ratings. Stable benchmark dataset. 1 million ratings from 6000 users on 4000 movies. Released 2/2003. README.txt ml-1m.zip (size: 6 MB, checksum) Permalink:

HETREC 2011

The 2nd International Workshop on Information Heterogeneity and Fusion in Recommender Systems (HetRec 2011, has released datasets from

Delicious, Last.fm

EVALUATING COLLABORATIVE FILTERING RECOMMENDER SYSTEMS 6 • J. L. Herlocker et al. the most successful technologies for recommender systems, called collabora- tive filtering, has been developed and improved over the past decade to the point where a wide variety of algorithms exist for generating recommenda- BEING ACCURATE IS NOT ENOUGH: HOW ACCURACY METRICS HAVE Being Accurate is Not Enough: How Accuracy Metrics have hurt Recommender Systems Sean M. McNee Abstract Recommender systems have shown great potential to COMBINING COLLABORATIVE FILTERING WITH PERSONAL AGENTS FOR Combining Collaborative Filtering with Personal Agents for Better Recommendations Nathaniel Good, J. Ben Schafer, Joseph A. Konstan, Al

Borchers,

DO YOU TRUST YOUR RECOMMENDATIONS? AN EXPLORATION OF 3 Model Builder (2) Predictor (4) Personal Information (1) Data Store (3) Predicted Preferences (5) Recommender (centralized or distributed) Fig.1. Conceptual model of the interaction between a user and a

recommender system.

WHAT IS GROUPLENS?

GroupLens is a research lab in the Department of Computer Science and Engineering at the University of Minnesota, Twin Cities specializing in recommender systems, online communities, mobile and ubiquitous technologies, digital libraries, and local geographic information systems. We advance the theory and practice of social computing by

building

MOVIELENS LATEST DATASETS MovieLens Latest Datasets. These datasets will change over time, and are not appropriate for reporting research results. We will keep the download links stable for automated downloads. We will not archive or make available previously released versions. Small: 100,000 ratings and 3,600 tag applications applied to 9,000 movies by 600 users. MOVIELENS 20M DATASET MovieLens 20M movie ratings. Stable benchmark dataset. 20 million ratings and 465,000 tag applications applied to 27,000 movies by 138,000 users. Includes tag genome data with 12 WIKILENS | GROUPLENS WikiLens was a generalized collaborative recommender system that allowed its community to define item types (e.g. beer) and categories (e.g. microbrews, pale ales, stouts), and then rate and get recommendations for items. It was taken offline in 2009 due to lack of system maintenance and support. This data set was extracted in

February 2008.

EACHMOVIE | GROUPLENS When EachMovie was shut down, the dataset was available to the public for use in research. MovieLens was originally based on this dataset. It contains 2,811,983 ratings entered by 72,916 for 1628 different movies, and it has been used in numerous CF publications. As of October, 2004, HP retired the EachMovie dataset.

BOOK-CROSSING

Book-Crossing. The BookCrossing (BX) dataset was collected by Cai-Nicolas Ziegler in a 4-week crawl (August / September 2004) from the Book-Crossing community with kind permission from Ron Hornbaker, CTO of Humankind Systems. It contains 278,858 users (anonymized but with demographic information) providing 1,149,780 ratings (explicit /

implicit

TAG GENOME DATA SET README Dataset. This data set contains the tag relevance values that make up the tag genome, described here. Tag relevance represents the relevance of a tag to a movie on a continuous scale from 0 to 1. Tag relevance values are provided for 9,734 movies and 1,128 tags. The data are contained in three files, tag_relevance.dat, movies.dat and tags.dat. SIMILARITY FUNCTIONS FOR USER-USER COLLABORATIVE FILTERING Similarity Functions for User-User Collaborative Filtering. By Michael Ekstrand on October 24, 2013. Typically, user-user collaborative filtering has used Pearson correlation to compare users. Early work tried Spearman correlation and (raw) cosine similarity, but found Pearson to work better, and the issue wasn’t revisited for quite

some time.

BEING ACCURATE IS NOT ENOUGH: HOW ACCURACY METRICS HAVE Being Accurate is Not Enough: How Accuracy Metrics have hurt Recommender Systems Sean M. McNee Abstract Recommender systems have shown great potential to DO YOU TRUST YOUR RECOMMENDATIONS? AN EXPLORATION OF 3 Model Builder (2) Predictor (4) Personal Information (1) Data Store (3) Predicted Preferences (5) Recommender (centralized or distributed) Fig.1. Conceptual model of the interaction between a user and a

recommender system.

GROUPLENSABOUTDATASETSPUBLICATIONSBLOGPROJECTSCONTACT US MovieLens is a web site that helps people find movies to watch. It has hundreds of thousands of registered users. We conduct online field experiments in MovieLens in the areas of automated content recommendation, recommendation interfaces, tagging-based recommenders and interfaces, member-maintained databases, and intelligent user

interface

MOVIELENS | GROUPLENSMOVIELENS 10M DATASETMOVIELENS 100K DATASETMOVIELENS TAG GENOME DATASET MovieLens 1B Synthetic Dataset. MovieLens 1B is a synthetic dataset that is expanded from the 20 million real-world ratings from ML-20M, distributed in support of MLPerf.Note that these data are distributed as .npz files, which you must read using python and numpy.. README DATASETS | GROUPLENS Datasets. GroupLens Research has collected and made available several datasets. Choose the one you’re interested in from the menu on the right. Before using these data sets, please review their README files for the usage licenses and other details. MOVIELENS 100K DATASET MovieLens 100K movie ratings. Stable benchmark dataset. 100,000 ratings from 1000 users on 1700 movies. Released 4/1998. README.txt

ml-100k.zip (size:

HETREC 2011

The 2nd International Workshop on Information Heterogeneity and Fusion in Recommender Systems (HetRec 2011, has released datasets from

Delicious, Last.fm

COMBINING COLLABORATIVE FILTERING WITH PERSONAL AGENTS FOR Combining Collaborative Filtering with Personal Agents for Better Recommendations Nathaniel Good, J. Ben Schafer, Joseph A. Konstan, Al

Borchers,

EXPLAINING COLLABORATIVE FILTERING RECOMMENDATIONSCOLLABORATIVE FILTERING ALGORITHMCOLLABORATIVE FILTERING MODELCOLLABORATIVE FILTERING PDFCOLLABORATIVE FILTERING TUTORIALDEEP COLLABORATIVE FILTERINGUSER BASED COLLABORATIVE FILTERING Explaining Collaborative Filtering Recommendations Jonathan L. Herlocker, Joseph A. Konstan, and John Riedl Dept. of Computer Science

and Engineering

TAGSPLANATIONS: EXPLAINING RECOMMENDATIONS USING TAGS We present a new type of explanation that uses tags as fea-tures, which we call a tagsplanation. The intermediary en-tity for tagsplanations is a tag or a set of tags. DO YOU TRUST YOUR RECOMMENDATIONS? AN EXPLORATION OF 3 Model Builder (2) Predictor (4) Personal Information (1) Data Store (3) Predicted Preferences (5) Recommender (centralized or distributed) Fig.1. Conceptual model of the interaction between a user and a

recommender system.

SHILLING RECOMMENDER SYSTEMS FOR FUN AND PROT Shilling Recommender Systems for Fun and Prot Shyong (Tony) K. Lam John Riedl GroupLens Research Computer Science and Engineering University of Minnesota GROUPLENSABOUTDATASETSPUBLICATIONSBLOGPROJECTSCONTACT US MovieLens is a web site that helps people find movies to watch. It has hundreds of thousands of registered users. We conduct online field experiments in MovieLens in the areas of automated content recommendation, recommendation interfaces, tagging-based recommenders and interfaces, member-maintained databases, and intelligent user

interface

MOVIELENS | GROUPLENSMOVIELENS 10M DATASETMOVIELENS 100K DATASETMOVIELENS TAG GENOME DATASET MovieLens 1B Synthetic Dataset. MovieLens 1B is a synthetic dataset that is expanded from the 20 million real-world ratings from ML-20M, distributed in support of MLPerf.Note that these data are distributed as .npz files, which you must read using python and numpy.. README DATASETS | GROUPLENS Datasets. GroupLens Research has collected and made available several datasets. Choose the one you’re interested in from the menu on the right. Before using these data sets, please review their README files for the usage licenses and other details. MOVIELENS 100K DATASET MovieLens 100K movie ratings. Stable benchmark dataset. 100,000 ratings from 1000 users on 1700 movies. Released 4/1998. README.txt

ml-100k.zip (size:

HETREC 2011

The 2nd International Workshop on Information Heterogeneity and Fusion in Recommender Systems (HetRec 2011, has released datasets from

Delicious, Last.fm

COMBINING COLLABORATIVE FILTERING WITH PERSONAL AGENTS FOR Combining Collaborative Filtering with Personal Agents for Better Recommendations Nathaniel Good, J. Ben Schafer, Joseph A. Konstan, Al

Borchers,

EXPLAINING COLLABORATIVE FILTERING RECOMMENDATIONSCOLLABORATIVE FILTERING ALGORITHMCOLLABORATIVE FILTERING MODELCOLLABORATIVE FILTERING PDFCOLLABORATIVE FILTERING TUTORIALDEEP COLLABORATIVE FILTERINGUSER BASED COLLABORATIVE FILTERING Explaining Collaborative Filtering Recommendations Jonathan L. Herlocker, Joseph A. Konstan, and John Riedl Dept. of Computer Science

and Engineering

TAGSPLANATIONS: EXPLAINING RECOMMENDATIONS USING TAGS We present a new type of explanation that uses tags as fea-tures, which we call a tagsplanation. The intermediary en-tity for tagsplanations is a tag or a set of tags. DO YOU TRUST YOUR RECOMMENDATIONS? AN EXPLORATION OF 3 Model Builder (2) Predictor (4) Personal Information (1) Data Store (3) Predicted Preferences (5) Recommender (centralized or distributed) Fig.1. Conceptual model of the interaction between a user and a

recommender system.

SHILLING RECOMMENDER SYSTEMS FOR FUN AND PROT Shilling Recommender Systems for Fun and Prot Shyong (Tony) K. Lam John Riedl GroupLens Research Computer Science and Engineering University of Minnesota

WHAT IS GROUPLENS?

GroupLens is a research lab in the Department of Computer Science and Engineering at the University of Minnesota, Twin Cities specializing in recommender systems, online communities, mobile and ubiquitous technologies, digital libraries, and local geographic information systems. We advance the theory and practice of social computing by

building

MOVIELENS LATEST DATASETS MovieLens Latest Datasets. These datasets will change over time, and are not appropriate for reporting research results. We will keep the download links stable for automated downloads. We will not archive or make available previously released versions. Small: 100,000 ratings and 3,600 tag applications applied to 9,000 movies by 600 users. MOVIELENS 1M DATASET MovieLens 1M movie ratings. Stable benchmark dataset. 1 million ratings from 6000 users on 4000 movies. Released 2/2003. README.txt ml-1m.zip (size: 6 MB, checksum) Permalink: MOVIELENS 20M DATASET MovieLens 20M movie ratings. Stable benchmark dataset. 20 million ratings and 465,000 tag applications applied to 27,000 movies by 138,000 users. Includes tag genome data with 12

BOOK-CROSSING

Book-Crossing. The BookCrossing (BX) dataset was collected by Cai-Nicolas Ziegler in a 4-week crawl (August / September 2004) from the Book-Crossing community with kind permission from Ron Hornbaker, CTO of Humankind Systems. It contains 278,858 users (anonymized but with demographic information) providing 1,149,780 ratings (explicit /

implicit

EVALUATING COLLABORATIVE FILTERING RECOMMENDER SYSTEMS 6 • J. L. Herlocker et al. the most successful technologies for recommender systems, called collabora- tive filtering, has been developed and improved over the past decade to the point where a wide variety of algorithms exist for generating recommenda- PREPARING FOR A GOOGLE TECHNICAL INTERVIEW Preparing for a Google technical interview. By Daniel Jarratt on October 14, 2014. GroupLens students and alumni successfully interview at Google on a regular basis. Several current GroupLens students have interned at the company, and our alumni have become Google research scientists and software engineers. I collected the following technical MODELING A DIALOGUE STRATEGY FOR PERSONALIZED MOVIE preferences. Each dialogue session required two subjects, one acting in the role of a movie recommender, and the other in the role of a client wanting movie recommendations. In order to avoid repetition of recommendation strategies in the IMPROVING RECOMMENDATION LISTS THROUGH TOPIC DIVERSIFICATION Improving Recommendation Lists Through Topic Diversification Cai-Nicolas Ziegler1∗ Sean M. McNee2 1 InstitutfurInformatik,Universit¨ atFreiburg¨ Georges-K¨ohler ACCURATE IS NOT ALWAYS GOOD: HOW ACCURACY METRICS HAVE 5 users, recommenders will continue to generate mismatched recommendations . In the end, recommenders exist to help

users. We, as

GROUPLENSABOUTDATASETSPUBLICATIONSBLOGPROJECTSCONTACT US Find bike routes that match the way you ride. Share your cycling knowledge with the community. Cyclopath is a geowiki: an editable map where anyone can share notes about roads and trails, enter tags about special locations, and fix map problems – like missing trails. MOVIELENS | GROUPLENSMOVIELENS 10M DATASETMOVIELENS 100K DATASETMOVIELENS TAG GENOME DATASET MovieLens 1B Synthetic Dataset. MovieLens 1B is a synthetic dataset that is expanded from the 20 million real-world ratings from ML-20M, distributed in support of MLPerf.Note that these data are distributed as .npz files, which you must read using python and numpy.. README DATASETS | GROUPLENS GroupLens Research has collected and made available several datasets. Choose the one you’re interested in from the menu on the right. Before using these data sets, please review their README files for the usage licenses and other details. MOVIELENS 100K DATASET MovieLens 100K movie ratings. Stable benchmark dataset. 100,000 ratings from 1000 users on 1700 movies. Released 4/1998. README.txt

ml-100k.zip (size:

PREPARING FOR A GOOGLE TECHNICAL INTERVIEW

HETREC 2011

The 2nd International Workshop on Information Heterogeneity and Fusion in Recommender Systems (HetRec 2011, has released datasets from

Delicious, Last.fm

COMBINING COLLABORATIVE FILTERING WITH PERSONAL AGENTS FOR Combining Collaborative Filtering with Personal Agents for Better Recommendations Nathaniel Good, J. Ben Schafer, Joseph A. Konstan, Al

Borchers,

EXPLAINING COLLABORATIVE FILTERING RECOMMENDATIONSCOLLABORATIVE FILTERING ALGORITHMCOLLABORATIVE FILTERING MODELCOLLABORATIVE FILTERING PDFCOLLABORATIVE FILTERING TUTORIALDEEP COLLABORATIVE FILTERINGUSER BASED COLLABORATIVE FILTERING Explaining Collaborative Filtering Recommendations Jonathan L. Herlocker, Joseph A. Konstan, and John Riedl Dept. of Computer Science

and Engineering

DO YOU TRUST YOUR RECOMMENDATIONS? AN EXPLORATION OF 3 Model Builder (2) Predictor (4) Personal Information (1) Data Store (3) Predicted Preferences (5) Recommender (centralized or distributed) Fig.1. Conceptual model of the interaction between a user and a

recommender system.

SHILLING RECOMMENDER SYSTEMS FOR FUN AND PROT Shilling Recommender Systems for Fun and Prot Shyong (Tony) K. Lam John Riedl GroupLens Research Computer Science and Engineering University of Minnesota GROUPLENSABOUTDATASETSPUBLICATIONSBLOGPROJECTSCONTACT US Find bike routes that match the way you ride. Share your cycling knowledge with the community. Cyclopath is a geowiki: an editable map where anyone can share notes about roads and trails, enter tags about special locations, and fix map problems – like missing trails. MOVIELENS | GROUPLENSMOVIELENS 10M DATASETMOVIELENS 100K DATASETMOVIELENS TAG GENOME DATASET MovieLens 1B Synthetic Dataset. MovieLens 1B is a synthetic dataset that is expanded from the 20 million real-world ratings from ML-20M, distributed in support of MLPerf.Note that these data are distributed as .npz files, which you must read using python and numpy.. README DATASETS | GROUPLENS GroupLens Research has collected and made available several datasets. Choose the one you’re interested in from the menu on the right. Before using these data sets, please review their README files for the usage licenses and other details. MOVIELENS 100K DATASET MovieLens 100K movie ratings. Stable benchmark dataset. 100,000 ratings from 1000 users on 1700 movies. Released 4/1998. README.txt

ml-100k.zip (size:

PREPARING FOR A GOOGLE TECHNICAL INTERVIEW

HETREC 2011

The 2nd International Workshop on Information Heterogeneity and Fusion in Recommender Systems (HetRec 2011, has released datasets from

Delicious, Last.fm

COMBINING COLLABORATIVE FILTERING WITH PERSONAL AGENTS FOR Combining Collaborative Filtering with Personal Agents for Better Recommendations Nathaniel Good, J. Ben Schafer, Joseph A. Konstan, Al

Borchers,

EXPLAINING COLLABORATIVE FILTERING RECOMMENDATIONSCOLLABORATIVE FILTERING ALGORITHMCOLLABORATIVE FILTERING MODELCOLLABORATIVE FILTERING PDFCOLLABORATIVE FILTERING TUTORIALDEEP COLLABORATIVE FILTERINGUSER BASED COLLABORATIVE FILTERING Explaining Collaborative Filtering Recommendations Jonathan L. Herlocker, Joseph A. Konstan, and John Riedl Dept. of Computer Science

and Engineering

DO YOU TRUST YOUR RECOMMENDATIONS? AN EXPLORATION OF 3 Model Builder (2) Predictor (4) Personal Information (1) Data Store (3) Predicted Preferences (5) Recommender (centralized or distributed) Fig.1. Conceptual model of the interaction between a user and a

recommender system.

SHILLING RECOMMENDER SYSTEMS FOR FUN AND PROT Shilling Recommender Systems for Fun and Prot Shyong (Tony) K. Lam John Riedl GroupLens Research Computer Science and Engineering University of Minnesota MOVIELENS LATEST DATASETS GroupLens gratefully acknowledges the support of the National Science Foundation under research grants IIS 05-34420, IIS 05-34692, IIS 03-24851, IIS 03-07459, CNS 02-24392, IIS 01-02229, IIS 99-78717, IIS 97-34442, DGE 95-54517, IIS 96-13960, IIS 94-10470, IIS 08-08692, BCS 07-29344, IIS 09-68483, IIS 10-17697, IIS 09-64695 and IIS 08-12148. MOVIELENS 1M DATASET MovieLens 1M movie ratings. Stable benchmark dataset. 1 million ratings from 6000 users on 4000 movies. Released 2/2003. README.txt ml-1m.zip (size: 6 MB, checksum) Permalink: MOVIELENS 20M DATASET MovieLens 20M movie ratings. Stable benchmark dataset. 20 million ratings and 465,000 tag applications applied to 27,000 movies by 138,000 users. Includes tag genome data with 12 PREPARING FOR A GOOGLE TECHNICAL INTERVIEW Read More from GroupLens Research. Tabletop Games in the Age of Remote Collaboration May 13, 2021; Human-Aided Information Retrieval to Create a Peer Support Group Meeting List July 21, 2020; System Design to Reduce Social Cost of Question Asking July 2, 2020; Designing Technology for Children: Review of Last Decade’s IDC Research June 22, 2020; Get More Citations by Keeping

BOOK-CROSSING

The BookCrossing (BX) dataset was collected by Cai-Nicolas Ziegler in a 4-week crawl (August / September 2004) from the Book-Crossing

community with

EVALUATING COLLABORATIVE FILTERING RECOMMENDER SYSTEMS 6 • J. L. Herlocker et al. the most successful technologies for recommender systems, called collabora- tive filtering, has been developed and improved over the past decade to the point where a wide variety of algorithms exist for generating recommenda- MODELING A DIALOGUE STRATEGY FOR PERSONALIZED MOVIE preferences. Each dialogue session required two subjects, one acting in the role of a movie recommender, and the other in the role of a client wanting movie recommendations. In order to avoid repetition of recommendation strategies in the IMPROVING RECOMMENDATION LISTS THROUGH TOPIC DIVERSIFICATION Improving Recommendation Lists Through Topic Diversification Cai-Nicolas Ziegler1∗ Sean M. McNee2 1 InstitutfurInformatik,Universit¨ atFreiburg¨ Georges-K¨ohler ACCURATE IS NOT ALWAYS GOOD: HOW ACCURACY METRICS HAVE 5 users, recommenders will continue to generate mismatched recommendations . In the end, recommenders exist to help

users. We, as

APPLICATION OF DIMENSIONALITY REDUCTION IN RECOMMENDER Application of Dimensionality Reduction in Recommender System -- A Case Study Badrul M. Sarwar, George Karypis, Joseph A. Konstan, John

T. Riedl

Toggle navigation

* about

* datasets

* publications

* blog

SOCIAL COMPUTING RESEARCH AT THE UNIVERSITY OF MINNESOTA GroupLens advances the theory and practice of social computing by building and understanding systems used by real people

FEATURED RESEARCH

We publish research articles in conferences and journals primarily in the field of computer science, but also in other fields including psychology, sociology, and medicine. See our blog for research highlights and our publications page for a comprehensive view of our research contributions. Here are excerpts from recent articles: LEARNING FROM LEARNING BUDDIES: OPPORTUNITIES FOR TECH TO CONNECT

ACROSS GENERATIONS

As a child, I spent a lot of time with my parents’ retired colleagues in our community who often helped take care of children like me when our working parents were occupied. Yet to say, not all children have such caring older adults when they grow up and many older adults don’t have younger generations in their community as they age. Today’s communities have programs that specifically aim to connect the two generations. These programs often seek older adults’ experience and expertise to support children’s growth. However, there are challenges that prevent older adults from benefiting in these programs and we see opportunities for technologies to address these challenges… See more VALUE SENSITIVE ALGORITHM DESIGN: METHOD, CASE STUDY AND LESSONS Intelligent algorithmic systems are assisting humans to make important decisions in a wide variety of critical domains. Examples include: helping judges decide whether defendants should be detained or released while awaiting trial; assisting child protection agencies in screening referral calls; and helping employers to filter job resumes. However, technically sound algorithms might fail in multiple ways. First, automation may worsen engagement with key users and stakeholders. For instance, a series of studies have shown that even when algorithmic predictions are proved to be more accurate than human predictions, domain experts and laypeople remain resistant to using the algorithms. Second, an approach that largely relies on automated processing of historical data might repeat and amplify historical stereotypes, discriminations, and prejudices… See more

FEATURED PROJECTS

We build and study real systems, going back to the release of MovieLens in 1997. See our projects page for a full list of active projects; see below for some featured projects. MovieLens is a web site that helps people find movies to watch. It has hundreds of thousands of registered users. We conduct online field experiments in MovieLens in the areas of automated content recommendation, recommendation interfaces, tagging-based recommenders and interfaces, member-maintained databases, and intelligent user

interface design.

Find bike routes that match the way you ride. Share your cycling knowledge with the community. Cyclopath is a geowiki: an editable map where anyone can share notes about roads and trails, enter tags about special locations, and fix map problems – like missing trails. Hundreds of Twin Cities cyclists are already doing this, making Cyclopath the most comprehensive and up-to-date bicycle information resource in the world. LensKit is an open source toolkit for building, researching, and studying recommender systems. Do you need a recommender for your next project? LensKit provides high-quality implementations of well-regarded collaborative filtering algorithms and is designed for integration into web applications and other similarly complex

environments.

MORE ABOUT GROUPLENS GroupLens is headed by faculty from the department of computer science and engineering at the University of Minnesota, and is home to a variety of students, staff, and visitors

.

JOE KONSTAN

LOREN TERVEEN

LANA YAROSH

STAY IN TOUCH

* Contact Us

* Follow us on Twitter

PROJECT LINKS

* MovieLens

* LensKit

* BookLens

* Cyclopath

CODE

* GroupLens on GitHub * GroupLens on Bitbucket GroupLens gratefully acknowledges the support of the National Science Foundation under research grants IIS 05-34420, IIS 05-34692, IIS 03-24851, IIS 03-07459, CNS 02-24392, IIS 01-02229, IIS 99-78717, IIS 97-34442, DGE 95-54517, IIS 96-13960, IIS 94-10470, IIS 08-08692, BCS 07-29344, IIS 09-68483, IIS 10-17697, IIS 09-64695 and IIS 08-12148.

� 2020 GroupLens