Supporting the education evidence portal via text mining

Area of Science:

• Information Science
• Education Technology

Background:

• The UK Education Evidence Portal (eep) aggregates content from 33 organizations, offering a centralized resource for the education community.
• The portal's extensive collection of over 500,000 documents presents challenges for users seeking specific information efficiently.
• Current search functionalities can yield numerous results, necessitating improved methods for users with limited time.

Purpose of the Study:

• To develop and demonstrate a prototype web interface integrating text-mining tools into the eep.
• To enhance the searching, browsing, and document-viewing experience for eep users.
• To facilitate more efficient information retrieval for education professionals.

Main Methods:

• Integration of text-mining tools into a prototype web interface for the eep.
• Development of automatic document classification based on a taxonomy.
• Implementation of automatic clustering of search results by document content similarity.
• Automatic identification and highlighting of key terms within documents.

Main Results:

• A prototype web interface demonstrating the integration of text-mining functionalities.
• Automatic classification of documents, enabling structured browsing.
• Clustering of search results, grouping similar documents for easier review.
• Automated identification and highlighting of key terms within search results.

Conclusions:

• Text-mining tools can significantly enhance the usability of large educational evidence portals.
• The ASSIST project's prototype demonstrates a viable approach to improving information access in education.
• These enhancements can revolutionize work practices by saving users time and improving the relevance of search results.