Relational Graph Embeddings for Table Retrieval

Trabelsi, Mohamed; Chen, Zhiyu; Davison, Brian D.; Heflin, Jeff

doi:10.1109/bigdata50022.2020.9378239

“…Shraga et al [37] use neural networks to learn unimodal features of a table which are combined into a multimodal representation. Tables can also be represented as graphs to solve table retrieval [9,41,46].…”

Section: Related Workmentioning

confidence: 99%

StruBERT: Structure-aware BERT for Table Search and Matching

Trabelsi,

Chen,

Zhang

et al. 2022

Preprint

Self Cite

0

View full text Add to dashboard Cite

A large amount of information is stored in data tables. Users can search for data tables using a keyword-based query. A table is composed primarily of data values that are organized in rows and columns providing implicit structural information. A table is usually accompanied by secondary information such as the caption, page title, etc., that form the textual information. Understanding the connection between the textual and structural information is an important yet neglected aspect in table retrieval as previous methods treat each source of information independently. In addition, users can search for data tables that are similar to an existing table, and this setting can be seen as a content-based table retrieval. In this paper, we propose StruBERT, a structure-aware BERT model that fuses the textual and structural information of a data table to produce context-aware representations for both textual and tabular content of a data table. StruBERT features are integrated in a new end-to-end neural ranking model to solve three table-related downstream tasks: keyword-and content-based table retrieval, and table similarity. We evaluate our approach using three datasets, and we demonstrate substantial improvements in terms of retrieval and classification metrics over state-of-the-art methods. CCS CONCEPTS• Information systems → Retrieval models and ranking; Structured text search.

show abstract

“…More recently, Xu et al (2020a) first construct a product knowledge graph and then propose a self-attention-enhanced distributed representation learning method with an efficient multi-task training schema to learn the graph embeddings, which can improve the performance of downstream tasks such as search ranking and recommendation. GNNs are also introduced in table retrieval, where Trabelsi et al (2020c) first construct a knowledge graph from table collections and then learn the graph embeddings in order to rank tables.…”

Section: Knowledge Graphsmentioning

confidence: 99%

“…Inspired by recent progress of transfer learning on graph neural networks, Trabelsi et al (2020c) proposed to represent a large collection of data tables using graphs. In particular, a knowledge graph representation using fact triples < subject, predicate, object> indicates the relations between entities, then R-GCN (Schlichtkrull et al, 2018), which is an extension of GCN for knowledge graphs, is applied on knowledge graphs to learn representations for graph nodes and relations.…”

Section: Structured Document Retrievalmentioning

confidence: 99%

Neural ranking models for document retrieval

Trabelsi

¹

,

Chen

²

,

Davison

³

et al. 2021

Self Cite

View full text Add to dashboard Cite

Ranking models are the main components of information retrieval systems. Several approaches to ranking are based on traditional machine learning algorithms using a set of hand-crafted features. Recently, researchers have leveraged deep learning models in information retrieval. These models are trained end-to-end to extract features from the raw data for ranking tasks, so that they overcome the limitations of hand-crafted features. A variety of deep learning models have been proposed, and each model presents a set of neural network components to extract features that are used for ranking. In this paper, we compare the proposed models in the literature along different dimensions in order to understand the major contributions and limitations of each model. In our discussion of the literature, we analyze the promising neural components, and propose future research directions. We also show the analogy between document retrieval and other retrieval tasks where the items to be ranked are structured documents, answers, images and videos.

show abstract

“…The Graph Convolutional Network (GCN) (Kipf and Welling 2017) can capture high order neighborhood information to learn representations of nodes in a graph. Inspired by recent progress of transfer learning on graph neural networks, a future research direction for ad-hoc table retrieval consists of representing a large collection of data tables using graphs (Trabelsi et al 2020c).…”

Section: 3mentioning

confidence: 99%

Neural ranking models for document retrieval

Trabelsi,

Chen,

Davison

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Ranking models are the main components of information retrieval systems. Several approaches to ranking are based on traditional machine learning algorithms using a set of hand-crafted features. Recently, researchers have leveraged deep learning models in information retrieval. These models are trained end-to-end to extract features from the raw data for ranking tasks, so that they overcome the limitations of hand-crafted features. A variety of deep learning models have been proposed, and each model presents a set of neural network components to extract features that are used for ranking. In this paper, we compare the proposed models in the literature along different dimensions in order to understand the major contributions and limitations of each model. In our discussion of the literature, we analyze the promising neural components, and propose future research directions. We also show the analogy between document retrieval and other retrieval tasks where the items to be ranked are structured documents, answers, images and videos.

show abstract

Relational Graph Embeddings for Table Retrieval

Cited by 4 publications

References 31 publications

StruBERT: Structure-aware BERT for Table Search and Matching

StruBERT: Structure-aware BERT for Table Search and Matching

Neural ranking models for document retrieval

Neural ranking models for document retrieval

Contact Info

Product

Resources

About