Pre-Trained Web Table Embeddings for Table Discovery

Gunther, Michael F.; Thiele, Maik; Gonsior, Julius; Lehner, Wolfgang

doi:10.1145/3464509.3464892

Cited by 6 publications

(7 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…With a fastText embedding model, we first get cell embeddings by averaging token embeddings in each cell and then aggregate cell embeddings to get a column embedding. More interestingly, web table embedding models [8] consider each cell as a single token (they concatenate tokens in a cell with underscores) and output embeddings at cell level. Nevertheless, we aggregate cell embeddings to derive the column embedding.…”

Section: Choices Of the Base Encodermentioning

confidence: 99%

“…Word embedding models have previously been used to find union-able tables. Two state-of-the-art choices are fastText and WTE (web table embeddings [8]). Language models have not thus far been used for the union-ability problem.…”

Section: Choices Of the Base Encodermentioning

confidence: 99%

“…[5] observed that only 13% of attribute values of their collected Open Data tables can be mapped to entities in YAGO [7], one of the largest and publicly available ontologies. Although word embeddings can provide more semantic coverage of attributes, they are subject to the training text corpus and may not generalize well to textual data in tables [8], [9]. Recent advance in tabular embeddings [10], [11] may mitigate the issue, however, we argue that these models in training do not take into account the indexing/search data structure, which is a core component of any efficient solution to table union search.…”

Section: Introductionmentioning

confidence: 97%

See 2 more Smart Citations

Pylon: Semantic Table Union Search in Data Lakes

Cong¹,

Nargesian²,

Jagadish³

2023

Preprint

View full text Add to dashboard Cite

The large size and fast growth of data repositories, such as data lakes, has spurred the need for data discovery to help analysts find related data. The problem has become challenging as (i) a user typically does not know what datasets exist in an enormous data repository; and (ii) there is usually a lack of a unified data model to capture the interrelationships between heterogeneous datasets from disparate sources. In this work, we address one important class of discovery needs: finding unionable tables.The task is to find tables in a data lake that can be unioned with a given query table. The challenge is to recognize unionable columns even if they are represented differently. In this paper, we propose a data-driven learning approach: specifically, an unsupervised representation learning and embedding retrieval task. Our key idea is to exploit self-supervised contrastive learning to learn an embedding model that takes into account the indexing/search data structure and produces embeddings close by for columns with semantically similar values while pushing apart columns with semantically dissimilar values. We then find union-able tables based on similarities between their constituent columns in embedding space. On a real-world data lake, we demonstrate that our best-performing model achieves significant improvements in precision (16% ↑), recall (17% ↑), and query response time (7x faster) compared to the state-of-the-art.

show abstract

Section: Choices Of the Base Encodermentioning

confidence: 99%

Section: Choices Of the Base Encodermentioning

confidence: 99%

Section: Introductionmentioning

confidence: 97%

See 1 more Smart Citation

Pylon: Semantic Table Union Search in Data Lakes

Cong¹,

Nargesian²,

Jagadish³

2023

Preprint

View full text Add to dashboard Cite

show abstract

“…A model is composed of a collection of real-valued vectors, each associated with a relevant DB term. The process of deriving vectors from DBderived text is called relational embedding, which is a very active area of research [2,6,14]. We provide initial vectors by textifying the database intelligently [3,19] and operate in two modes: assigning vectors to either whole tuples or to columns (per-tuple).…”

Section: Relational Embeddingmentioning

confidence: 99%

Towards practical approximate lineage

Leybovich

Shmueli

2022

Proceedings of the 14th International Workshop on the Theory and Practice of Provenance

View full text Add to dashboard Cite

Traditionally, provenance and lineage mainly referred to query results. We take a more holistic approach. We consider a system in which tuples (records) that are produced by a query may affect other tuple insertions into the DB, as part of a normal workflow. Therefore, we consider both direct lineage (dependence of a query result on database tuples directly used in solving the query) and distant lineage (dependence on older tuples that caused the existence

show abstract

“…Such a model is composed of a collection of real-valued vectors, each associated with a relevant DB term. The process of deriving vectors from DB-derived text is called relational embedding, which is a very active area of research [2,8,20].…”

Section: Word Vectors Modelmentioning

confidence: 99%

ML Based Lineage in Databases

Leybovich,

Shmueli

2021

Preprint

View full text Add to dashboard Cite

We track the lineage of tuples throughout their database lifetime. That is, we consider a scenario in which tuples (records) that are produced by a query may affect other tuple insertions into the DB, as part of a normal workflow. As time goes on, exact provenance explanations for such tuples become deeply nested, increasingly consuming space, and resulting in decreased clarity and readability.We present a novel approach for approximating lineage tracking, using a Machine Learning (ML) and Natural Language Processing (NLP) technique; namely, word embedding. The basic idea is summarizing (and approximating) the lineage of each tuple via a small set of constant-size vectors (the number of vectors per-tuple is a hyperparameter). For explicitly (and independently of DB contents) inserted tuples -the vectors are obtained via a pre-trained word vectors model over their underlying database domain "text". During the execution of a query, we construct the lineage vectors of the final (and intermediate) result tuples in a similar fashion to that of semiring-based exact provenance calculations. We extend the + and • operations to generate sets of lineage vectors, while retaining the ability to propagate information and preserve the compact representation. Therefore, our solution does not suffer from space complexity blow-up over time, and it "naturally ranks" explanations to the existence of a tuple.We devise a genetics-inspired improvement to our basic method. The data columns of an entity (and potentially other columns) are a tuple's basic properties, i.e., the "genes" that combine to form its genetic code. We design an alternative lineage tracking mechanism, that of keeping track of and querying lineage (via embeddings) at the column ("gene") level; thereby, we manage to better distinguish between the provenance features and the textual characteristics of a tuple. Finding the lineage of a tuple in the DB is analogous to finding its predecessors via DNA examination.We further introduce several improvements and extensions to the basic method: tuple creation timestamp, column emphasis and query dependency DAG.We integrate our lineage computations into the PostgreSQL system via an extension (ProvSQL) and extensive experiments exhibit useful results in terms of accuracy against exact, semiring-based, justifications, especially for the column-based (CV) method which exhibits high precision and high per-level recall. In the experiments, we focus on tuples with multiple generations of tuples in their lifelong lineage and analyze them in terms of direct and distant lineage.

show abstract

Pre-Trained Web Table Embeddings for Table Discovery

Cited by 6 publications

References 28 publications

Pylon: Semantic Table Union Search in Data Lakes

Pylon: Semantic Table Union Search in Data Lakes

Towards practical approximate lineage

ML Based Lineage in Databases

Contact Info

Product

Resources

About