The mEducator Linked Educational Resources dataset has been created as part of the mEducator project (2009-2012, http://www.meducator.net) with the aim to implement and critically evaluate existing standards and reference models in the field of e-learning in order to enable specialized state-of-the-art medical educational content to be discovered, retrieved, shared and re-used across European higher academic institutions (Dietze et al., 2013). The approach consists of two main steps: (1) semantic integration, which involves the integration of educational resources and data into a coherent data graph and (2) social integration, which involves the incorporation of the educational data graph into a heterogeneous social network, composed of educational resources and humans.
For the particular implementation in the domain of health sciences education, a unified mEducator RDF schema (Mitsopoulou et al., 2011) was developed, to address specifically medical educational resources. Metadata have been automatically enriched with additional information to ensure disambiguation, as well as to allow correlation with similar or related information and resources already available in the Linked Data cloud of datasets, ranging from general-purpose ones to domain-specific educational datasets.
The mEducator weighted graph provides a means of clustering resources originating from distinct repositories, with the aim of producing new connections between similar resources that were not explicitly connected before.
The educational resources dataset, as of September 2012, exhibits 375 resources with the following overall characteristics:
- educational level of resources: 25% undergraduate, 17% postgraduate, 19%continuing education, 19% patient-public education, 4% other;
- language of educational resources: more than 15 languages, 43% English, 18%Bulgarian, 7% French, 6% Greek, 2% Romanian, 2% Italian, 2% Finish;
- resource types: conventional (text, presentation, notes) 49%, multimedia 16%, case/problem based material 12%, web 2.0 4%, serious games 2%;used by educators and learners from more than 20 different countries, from various related fields: 90% medical, 5% medical informatics, 4% biology, 1% medical physics.
MetaMorphosis+ TOOL
The data and services integration APIs and datasets are fully integrated in the MetaMorphosis+ (http://metamorphosis.med.duth.gr/) environment, which merges the paradigms of semantic and social web to produce an environment for sharing linked educational resources.
MetaMorphosis+ is implemented using the Elgg open source social engine (http://www.elgg.org/) appropriately modified to allow for the formation of a heterogeneous human/resources network as well as the exploitation of the underlying semantic framework.
The educational resource framework is built based on connections and shared annotations between different educational resources, as these are created via the automatic and/or semi-automatic metadata enrichment mechanism developed in the project. The user entered keyword is used to retrieve all relevant standardized concepts from the medical ontologies/vocabularies available via NCBO BioPortal (https://bioportal.bioontology.org/), using either the entire BioPortal collection or only specific vocabularies specified by the user.
MetaMorphosis+ is used by a total of 560 registered users (as of September 25, 2011). Almost half of the resources (43%) are in English, while there is a representation of more than 15 other European languages. The majority of resources (80%) have metadata described in English language, while the rest have metadata in other languages. The resources included in the environment are distributed among the various educational levels, 25% intended for undergraduate medical education, 17% intended for postgraduate/resident studies and 19% for continuing life-long education, while 19% are intended for educating the public.
EVALUATION and EDUCATIONAL IMPLICATION
When an educator sets the context and goals of a new educational experience, he/she will overview existing content and/or search for new relative content and then repurpose and re-organize content to fit the purpose of the new educational experience. In MetaMorphosis+ repurposing history and inheritance are used as basic social relationship among educational resources in order to cluster resources into families. Clustering of resources has been evaluated from two perspectives: potential for interlinking and support for exploratory search.
The resources in the dataset were classified indipendently by two subjects: a computer scientist and a medical doctor. Based on the available metadata, they were asked to group the dataset first in 5 clusters, then in 10 clusters and in 15 clusters. A standard K-means was run at 5, 10, and 15, using the following metadata fields: Title, Description, Discipline, DisciplineSpecialty, and Keywords.
The results show a good coherence of the obtained clusters, since the average F-measure ranges from 0.74 (5 clusters) to 0.55 (15 clusters). Typical F-measures obtained by k-means are in the range 0.12-0.13 for the Citeseer dataset and 0.42-0.44 for the PIMO (Personal Information Model from a sematic desktop system) dataset [Grimnes, 08]. The number of resources tends to stay stable across different classifications, whereas the finer grained classification of 10 or 15 clusters appears suitable to discover potential interlinks across repositories.
Although scientific discovery is not the main purpose of the system, the implied educational implication is clear, since cluster analysis can stimulate the learners in generating questions, hypothesis and in the making of conceptual connections.
REFERENCES
Dietze et al. (2013) “Socio-semantic Integration of Educational Resources - the Case of the mEducator Project”. Journal of Universal Computer Science, vol. 19, no. 11 (2013), 1543-1569 submitted: 15/12/12, accepted: 28/5/13, appeared: 1/6/13 © J.UCS
Grimnes, G. A., Edwards, P., Preece, A. (2008). Instance based clustering of semantic web resources. In S. Bechhofer et al. (Eds). The semantic web: research and applications. Proceedings of the 5th European semantic web conference. Tenerife, Canary Islands, Spain (pp. 303-317). LNCS 502: Springer-Verlag, 2008.
Mitsopoulo, .E, Taibi, D., Giordano, D., Dietze, S., Yu, H. Q., Bamidis, P., Bratsas, C., Woodham, L. (2011). Connecting medical educational resources to the Linked Data cloud: the mEducator RDF Schema, store and API, in the Proceedings of Linked Learning 2011: 1st Int. Workshop on eLearning Approaches for the Linked Data Age, Herakleio, Greece, 2011.