This repository contains the implementation of my master's thesis, which focuses on computing PCA-based validation scores for knowledge graphs. The goal is to calculate PCA_valid and PCA_invalid scores for entities in the KG by validating the graph against a set of SHACL-like constraints. The pipeline involves:
1- Validating the knowledge graph using provided constraint rules (in .ttl format).
2- Enriching the KG with validation results (valid or invalid).
3- Calculating PCA scores based on the enriched graph.
4- Using these scores in symbolic prediction tasks.