Abstract

This project leverages OpenSearch to implement scalable vector and k-nearest neighbors (k-NN) search across multiple indices. The indices house learning opportunities and ESCO skills, enabling efficient retrieval of relevant content based on user queries and preferences. Various configurations of the indices were explored during the prototype phase to optimize search performance and relevance. This approach demonstrates the potential of vector-based search for enhancing educational content discovery, paving the way for improved user engagement and tailored learning experiences.