New tools and technologies for better access to museum collections: CATCHPlus results

The main purpose of CATCHPlus (Continuous Access to Cultural Heritage Plus) is to build usable tools and services for the entire (Dutch) heritage sector.
This software leads to better disclosure and larger accessibility of collections from heritage institutions. The unique cooperation between large heritage institutions (museums, archives and libraries), universities and creative sector companies in CATCHPlus, creates a new crossroad of IT and cultural heritage.
The products from CATCHPlus promote cooperation and coordination in the information infrastructure of the heritage sector. The results are all developed in Open Access and therefore available to everyone interested, which is especially interesting for smaller museums and institutions that cannot develop their own tools.

The program is finishing end of Juny 2012 and therefore more and more results are becoming available. They range from an Art-Recommending tool (CHIP-API Rijksmuseum and Amsterdam Museum) based on user profiles stored in a repository; a tool for automatic speech recognition (CHORALPLUS) to search in audio files; a search machine for melody-recognition (Witchcraft, Meertens Institute); tools for smart description MuseumPlus (City museum The Hague); for manuscript recognition (Scratch4all, National Archives) to get access to scanned handwritten information; a Documentalist Support System (Institute of Sound and Vision) and combined and aligned vocabularies (National Library of the Netherlands).

This paper will zoom in on the Art-Recommender (CHIPAPI) that was developed in cooperation with the Rijksmuseum and Amsterdam Museum. The tool makes use of the vast Collection of the Amsterdam Museum.

CHIPAPI consists of two elements: a rating tool to determine personal favorites by enabling users to evaluate digital objects from the collection . These personal favorites will be stored in a  User Profile Repository. The Art Recommender can use metadata-descriptions of a selected set of objects and information from the User Profile Repository to provide recommendations according to the user’s prefernces. In the user profile, user-id’s will be authenticated and linked to personal information, favorite objects and concepts. This user profile is set up in a generic way so that it can serve a more general purpose for other heritage institutions as well. The idea is that user will provide their profile data to different services enabling them to provide personalised solutions and suggestions. Visitors of the website of the Amsterdam Museum, as well as visitors of the museum can get a personalized tour of the collection. In order to be able to provide such a service their user profiles need to be mapped first.

The paper will focus on the added value and generic application of the art-recommender.

It will also deal with some questions: do people actually want these suggestions? How are they using it and how about its predictability of the results?