Research

The steve team concluded work on its two-year IMLS National Leadership Grant, “Researching Social Tagging and Folksonomy in the Art Museum” in December, 2008. With two years of productive research completed, the group shifted its focus from pure research to facilitating the widespread acceptance of tagging as a tool for museums, and the broad deployment of the steve tagging tools in museums and other cultural heritage organizations. However, research will continue to play a key role in project activities, and we hope to build on the research infrastructure that our data collection environment—the steve tagger—provides for the ongoing collection of robust data about taggers and their behaviors and attitudes. We are currently participating in a 3-year grant that considers of text mining, tagging, and text inferencing, “T3: Text, Tags, Trust,” and also hope to cultivate relationships with researchers who share our interest in social tagging, as well as with programmers who are interested in building the tools that will be needed to continue our work with term analysis, data processing, and disambiguation. Other research interests include the study of the effectiveness of guided or faceted tagging; the collection of tags from expert communities; and the collection and creation of multilingual tags.

Current Research Grant: T3

Our current collaboration with the University of Maryland and the Indianapolis Museum of Art is a research effort that combines text mining, social tagging, and trust inferencing techniques to enrich metadata and personalize retrieval. The project, “T3: Text, Terms, Trust,” is a 3-year National Leadership Grant for Research funded in2008 by the U.S. Institute for Museum and Library Services.

T3 is a collaborative, cross-disciplinary project comprised of academic researchers, digital librarians, and museum professionals. In the research, the project team will explore the application of techniques from computational linguistics and social tagging to the creation of links between the formal academic language of museums and the vernacular language of social tagging. We will take advantage of text mining algorithms, taxonomies, and lexical resources to understand, disambiguate, and/or enhance terms; to aid users in tagging images; and to retrieve images based on tags assigned from many different perspectives. Understanding the trust a user places in particular metadata sources may be used to infer a weighted set of results for search. It is hoped that developing new types of ranking algorithms, and considering term relationships from lexical resources, will produce high-quality, focused, personalized retrieval of works from museum collections.

A working group of museum professionals will inform the research, ensuring that the tools developed to support the research have immediate usefulness in “live” deployments of the steve tagger and other tagging environments.

2006 Research Grant Results

We have recently completed a 2-year National LeadershipGrant for Research funded by the U.S. Institute of Museum and LibraryServices, “Researching Social Tagging and Folksonomy in the ArtMuseum.” The final results of our work validate our key hypothesis:that social tagging can enhance access to museum collections by addinguseful terms to existing museum documentation. In all, more than 90,000tags, describing nearly 1,750 artworks and supplied by 2,275 users,were collected. The team has issued two documents that document thatproject and its findings: a project report, and a detailed analysis ofthe project data authored by Jennifer Trant, the project’s PrincipalInvestigator. Download the research analysis and results (10 MB) or the final project report (5 MB) here.

Data from the 2006 Research Project

The steve team has made the raw data collected during the 2006-08 research project available to scholars and students who may be interested in studying it to answer a range of questions about social tagging, language, cognition, and other matters. The data is available as a downloadable zip archive that includes tags, object metadata, and thumbnail images. Downloading the data signifies agreement to terms of the end user license, which can be viewed here.

In addition to those researchers who may be interested in studying the data to answer research questions of their own, students and others may be interested in working with the data set to consider questions about social tagging and museum collections raised during the research work, but not, ultimately, answered (either due to limitations of time and resources or because we believed that the data might not support our queries). We are currently working with colleagues at the University of California at Los Angeles to develop a brief list of questions that students or other researchers may find interesting, and will post it here shortly.

Comments are closed.