The Hofmethode: Computing Semantic Similarities between E-Learning Products

Oliver Michel, Damian Läge

Abstract


The key task in building useful e-learning repositories is to develop a system with an algorithm allowing users to retrieve information that corresponds to their specific requirements. To achieve this, products (or their verbal descriptions, i.e. presented in metadata) need to be compared and structured according to the results of this comparison. Such structuring is crucial insofar as there are many search results that correspond to the entered keyword. The Hofmethode is an algorithm (based on psychological considerations) to compute semantic similarities between texts and therefore offer a way to compare e-learning products. The computed similarity values are used to build semantic maps in which the products are visually arranged according to their similarities. The paper describes how the Hofmethode is implemented in the online database edulap, and how it contributes to help the user to explore the data in which he is interested.

Keywords


e-learning; information retrieval; machine text understanding

Full Text:

PDF



International Journal of Emerging Technologies in Learning. ISSN: 1863-0383
Creative Commons License SPARC Europe Seal
Indexing:
Engineering Information logo INSPEC logo DBLP logo ELSEVIER Scopus logo EDiTLib logo EBSCO logo Ulrich's logo DOAJ logo Google Scholar logo Microsoft® Academic Search