About this document
Joinable Table Discovery in Data Lakes by Himanshu Singhal is a document available to read on EtoBox.
The paper presents PEXESO, a framework for efficiently discovering joinable tables in data lakes using high-dimensional similarity. Traditional equi-join methods are inadequate due to issues like misspellings and varying formats, while PEXESO employs a block-and-verify strategy with pivot-based filtering to enhance joinability identification. Experimental results demonstrate that PEXESO significantly outperforms existing methods in both precision and recall, making it valuable for data enrichment in machine
- Author
- Himanshu Singhal
- Language
- EN