Graph Neural Network Approach to Semantic Type Detection in Tables

Ehsan Hoseinzade, Ke Wang
4/30/2024
cs.LGcs.CL

Official code: hoseinzadeehsan/GAIT (5 stars, MIT, last commit 2024-05-04).

Abstract

This study addresses the challenge of detecting semantic column types in relational tables, a key task in many real-world applications. While language models like BERT have improved prediction accuracy, their token input constraints limit the simultaneous processing of intra-table and inter-table information. We propose a novel approach using Graph Neural Networks (GNNs) to model intra-table dependencies, allowing language models to focus on inter-table information. Our proposed method not only outperforms existing state-of-the-art algorithms but also offers novel insights into the utility and functionality of various GNN types for semantic type detection. The code is available at https://github.com/hoseinzadeehsan/GAIT

View on arXivView PDF

Code Implementations(1)

hoseinzadeehsan/GAITOfficial100%
MIT

Graph Neural Network Approach to Semantic Type Detection in Tables

50PythonFeb 6, 20242 years agoMIT

Similar Papers

Comprehending Semantic Types in JSON Data with Graph Neural Networks

85% match2023

Sherlock: A Deep Learning Approach to Semantic Data Type Detection

82% match1 repo1912019

Tree Matching Networks for Natural Language Inference: Parameter-Efficient Semantic Understanding via Dependency Parse Trees

79% match2025

GNN: Graph Neural Network and Large Language Model for Data Discovery

78% match2024

Refining Interactions: Enhancing Anisotropy in Graph Neural Networks with Language Semantics

78% match2025

Boosting Relational Deep Learning with Pretrained Tabular Models

78% match1 repo22025

Cite this paper

@article{hoseinzade2024graph,
  title  = {Graph Neural Network Approach to Semantic Type Detection in Tables},
  author = {Ehsan Hoseinzade and Ke Wang},
  year   = {2024},
  eprint = {2405.00123},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2405.00123v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2405.00123.svg)](https://codeofpaper.com/paper/2405.00123)

Discussion