Bringing Database Systems into Machine Learning Workflows

Invited talk by Yiming Wu (DBIS Uni Basel) on bringing Database Systems into Machine Learning Workflows

Machine learning and database systems are two foundational concepts in modern data science. Database systems provide efficient and reliable methods for storing, managing, and analyzing large-scale data, while machine learning techniques enable the extraction of valuable insights from this data. 
In this talk, we will demonstrate the benefits of integrating database systems into machine learning workflows, using PyDuckPGQ as an example. PyDuckPGQ simplifies the retrieval of relational data as graph objects that can be used in machine learning workflows. This integration not only bridges the gap between property graph databases and machine learning frameworks but also opens up new possibilities for data modeling and analysis.
Evaluation results on social network benchmark datasets of the Linked Data Benchmark Council will be discussed to show the functionality of the integration and the effectiveness of our optimizations.

Veranstaltung übernehmen als iCal