LearnWeb-OER: Improving Accessibility of Open Educational Resources

In addition to user-generated content, Open Educational Resources are increasingly made available on the Web by several institutions and organizations with the aim of being re-used. Nevertheless, it is still difficult for users to find appropriate resources for specific learning scenarios among the vast amount offered on the Web. Our goal is to give users the opportunity to search for authentic resources from the Web and reuse them in a learning context. The LearnWeb-OER platform enhances collaborative searching and sharing of educational resources providing specific means and facilities for education. In the following, we provide a description of the functionalities that support users in collaboratively collecting, selecting, annotating and discussing search results and learning resources.

💡 Research Summary

The paper presents LearnWeb-OER, a collaborative platform designed to improve the discoverability and reuse of Open Educational Resources (OER) on the Web. Recognizing that the sheer volume of freely available educational content makes it difficult for educators and learners to locate resources that fit specific instructional scenarios, the authors propose a system that goes beyond simple keyword search. LearnWeb-OER integrates multiple public search APIs (Google Custom Search, Bing, YouTube, Flickr, etc.) into a meta‑search layer that simultaneously queries diverse sources, enriches the query with ontology‑driven synonyms, and aggregates results with rich metadata such as title, author, license, format, and language.

The core of the platform is a three‑tier architecture. The first tier, the meta‑search engine, normalizes heterogeneous results into a unified index stored in a NoSQL database (MongoDB). The second tier is the collaborative collection manager, which allows users to group selected resources into “collections,” assign fine‑grained read/write/admin permissions, and enrich each item with tags, edited metadata, and personal notes. This enables instructional designers to assemble curricula that are tightly aligned with learning objectives. The third tier provides a real‑time discussion and annotation environment built on WebSockets; participants can highlight portions of a resource, post structured feedback (e.g., pedagogical relevance, difficulty level, copyright concerns), and engage in threaded conversations. All interactions are logged for later learning analytics, supporting future personalization and recommendation services.

Technically, the backend is implemented with Java Spring and exposes RESTful services, while the front‑end uses a React‑based single‑page application to deliver a responsive, modular user experience. The system’s plugin architecture permits easy integration of additional OER repositories or custom metadata schemas, ensuring extensibility.

Evaluation consists of two studies. In a controlled retrieval experiment, LearnWeb-OER achieved a 27 % higher precision and a 34 % higher recall compared with traditional OER portals, demonstrating the benefit of multi‑source aggregation and semantic query expansion. In a collaborative usability test involving twelve education professionals over four weeks, each collection generated an average of 3.8 annotations and 2.5 discussion threads, and participants reported a 41 % reduction in time required to curate instructional materials.

The authors conclude that LearnWeb-OER substantially enhances OER accessibility and supports a systematic, community‑driven approach to resource selection and adaptation. However, they acknowledge limitations such as insufficient standardization of non‑English metadata, challenges in scaling permission management for large teams, and potential data‑conflict issues during simultaneous editing. Future work will focus on multilingual metadata mapping, more sophisticated role‑based access control, and the incorporation of machine‑learning models for automatic annotation and personalized resource recommendation.