Model Join

Model Join is a novel framework designed to perform data analytics on joined tables without direct access to the original raw data, leveraging pre-trained models of individual tables instead. Current research focuses on developing efficient model architectures and algorithms to approximate the results of joining large datasets, addressing challenges in accuracy and scalability, and exploring the use of models like BERT and ViT. This approach offers significant advantages for privacy-sensitive data, reduces computational costs associated with managing massive datasets, and enables knowledge discovery tasks on data that may be unavailable or impractical to directly join.