Knowledge Base and RAG Evaluation Dataset

This dataset consists of two tabs: 1. Knowledge Base: This tab contains key terms (hypernyms), their associated skills (hyponyms), definitions and KSA-O classifications. It follows the structure of the dataset "Automotive Industry Skills Taxonomy" . 2. Testing Set: This tab is used to evaluate a RAG (Retrieval-Augmented Generation) model that has access to the knowledge base from the first tab. Structure of the Testing Set (Tab 2): * Text: A hyponym (variation) of a skill. * Label: Categorizes the text into one of three unique values based on its presence in the knowledge base:     -New hypernym: A key term that does not exist in the knowledge base.      -Existing hyponym : A variation of a skill that already exists in the knowledge base.      -New hyponym: A new variation that can be mapped into an existing hypernym. * Original_key: Original hypernym associated with the hyponyms in the "Text" column. The terms in this column correspond to existing pairings in the taxonomy from "Automotive Industry Skills Taxonomy".