CHORUS: Foundation Models for Unified Data Discovery and Exploration

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Kayali, Moe, Lykov, Anton, Fountalis, Ilias, Vasiloglou, Nikolaos, Olteanu, Dan, Suciu, Dan
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929303900389376
author Kayali, Moe
Lykov, Anton
Fountalis, Ilias
Vasiloglou, Nikolaos
Olteanu, Dan
Suciu, Dan
author_facet Kayali, Moe
Lykov, Anton
Fountalis, Ilias
Vasiloglou, Nikolaos
Olteanu, Dan
Suciu, Dan
contents We apply foundation models to data discovery and exploration tasks. Foundation models include large language models (LLMs) that show promising performance on a range of diverse tasks unrelated to their training. We show that these models are highly applicable to the data discovery and data exploration domain. When carefully used, they have superior capability on three representative tasks: table-class detection, column-type annotation and join-column prediction. On all three tasks, we show that a foundation-model-based approach outperforms the task-specific models and so the state of the art. Further, our approach often surpasses human-expert task performance. We investigate the fundamental characteristics of this approach including generalizability to several foundation models and the impact of non-determinism on the outputs. All in all, this suggests a future direction in which disparate data management tasks can be unified under foundation models.
format Preprint
id arxiv_https___arxiv_org_abs_2306_09610
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle CHORUS: Foundation Models for Unified Data Discovery and Exploration
Kayali, Moe
Lykov, Anton
Fountalis, Ilias
Vasiloglou, Nikolaos
Olteanu, Dan
Suciu, Dan
Databases
Machine Learning
We apply foundation models to data discovery and exploration tasks. Foundation models include large language models (LLMs) that show promising performance on a range of diverse tasks unrelated to their training. We show that these models are highly applicable to the data discovery and data exploration domain. When carefully used, they have superior capability on three representative tasks: table-class detection, column-type annotation and join-column prediction. On all three tasks, we show that a foundation-model-based approach outperforms the task-specific models and so the state of the art. Further, our approach often surpasses human-expert task performance. We investigate the fundamental characteristics of this approach including generalizability to several foundation models and the impact of non-determinism on the outputs. All in all, this suggests a future direction in which disparate data management tasks can be unified under foundation models.
title CHORUS: Foundation Models for Unified Data Discovery and Exploration
topic Databases
Machine Learning
url https://arxiv.org/abs/2306.09610