Interoperable subject vocabulary service
Download the full report
This project addressed the perennial problem of the accuracy of subject searching in course information websites. Currently learners have to grapple with different search mechanics in otherwise very similar systems with little certainty that results obtained will relate well to the search terms entered. The project developed a pilot version of a subject vocabulary mapping service on the internet for course search purposes.
Executive Summary
The Advanced Search Algorithm Protocol (ASAP) prototype provides a ranked list of courses relevant to a user-selected keyword, using a specially designed algorithm and a mapping to a highly granular ‘spine’ thesaurus to give superior results.
The prototype is designed to show the efficacy of the search method, which uses a bespoke thesaurus to match search keywords and phrases. As course files are imported into the ASAP system, courses are mapped onto the thesaurus using many classification systems and keyword lists, so that classification is more accurate than conventional methods. By using a number of different classification methods, ASAP provides both a common, generic approach to mapping courses onto the thesaurus and a tailored approach for institution and even course specific mappings. Searches are conducted via an extensive but controlled list of terms used as an entry point into the thesaurus. The system then tracks the relationships through the thesaurus to build a results set containing the course data, relationships and scores logged against relevance criteria. The results are presented to the user in a variety of ways to demonstrate the flexibility of this approach.
ASAP prototype
ASAP script (Word)
The quality of search results was assessed heuristically against two leading course search websites and against Google Desktop search. ASAP searching provided a higher proportion of good quality matches and a prioritised list unequalled across all the comparator search techniques.
The project made highly successful use of agile prototyping development techniques to develop the service. Thesaurus development and mapping were carried out using MultiTes software. Data was obtained via XCRi-CAP 1.1 feeds, concentrating on a data set from the Open University.
The project team was pleased to consult with both the JISC High Level Thesaurus project and the Information Standards Board Vocabulary Management Service and obtained useful feedback from both on our outputs and potential outcomes.
In addition to the successful ASAP prototype service, the project was able to start the creation of the XCRi eXchange Platform (XXP), an XCRi-CAP enabled service and application platform, upon which many more services can be built. The project team members are committed to making sustained progress with both ASAP and the XXP platform. An important outcome of this project was simply to show that when course advertising data is published in the XCRi-CAP standard format, then new useful services can be developed and delivered.