SparkKG-ML: A Library to Facilitate End-to-End Large-Scale Machine Learning Over Knowledge Graphs in Python

Download PDF
Open Webpage

Bedirhan Gergin, Charalampos Chelmis

Published: 2024, Last Modified: 27 Jan 2026ISWC (3) 2024EveryoneRevisionsBibTeXCC BY-SA 4.0
Abstract: This paper presents SparkKG–ML, the first open–source library for Machine Learning at scale over semantic data stored in Knowledge Graphs directly in Python. SparkKG–ML serves as a bridge between (i) the Semantic Web data model, (ii) the distributed computing capabilities of Apache Spark, and (iii) the Python ecosystem. By harnessing the flexibility of Python and the scalability of Spark, SparkKG–ML reduces the barriers for Data Scientists and Machine Learning researchers to work with semantic data, and for Semantic Web experts to develop Machine Learning models.