TAIPAN: Automatic Property Mapping for Tabular Data

Ermilov, Ivan; Ngomo, Axel-Cyrille Ngonga

doi:10.1007/978-3-319-49004-5_11

Cited by 19 publications

(26 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In order to perform KG completion from web tables, we must first align their structure and content with the KG, a problem broadly referred to as table interpretation. Table interpretation has been the subject of several prior works [16,27,33,17,18,28,9,12]. Similar to our research, these works primarily focus on the interpretation of entity tables, i.e., tables where each row describes one entity and columns represent attributes.…”

Section: Introductionmentioning

confidence: 80%

Extracting Novel Facts from Tables for Knowledge Graph Completion

Kruit

Boncz

Urbani

2019

Lecture Notes in Computer Science

View full text Add to dashboard Cite

We propose a new end-to-end method for extending a Knowledge Graph (KG) from tables. Existing techniques tend to interpret tables by focusing on information that is already in the KG, and therefore tend to extract many redundant facts. Our method aims to find more novel facts. We introduce a new technique for table interpretation based on a scalable graphical model using entity similarities. Our method further disambiguates cell values using KG embeddings as additional ranking method. Other distinctive features are the lack of assumptions about the underlying KG and the enabling of a fine-grained tuning of the precision/recall trade-off of extracted facts. Our experiments show that our approach has a higher recall during the interpretation process than the state-of-the-art, and is more resistant against the bias observed in extracting mostly redundant facts since it produces more novel extractions.

show abstract

Section: Introductionmentioning

confidence: 80%

Extracting Novel Facts from Tables for Knowledge Graph Completion

Kruit

Boncz

Urbani

2019

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

“…One column per table is a Subject Column. That is, a textual column which represents the main subject of the table and connects the other columns semantically through binary relations [6,7,10]. Those connections are represented through properties from a KB.…”

Section: Definitionsmentioning

confidence: 99%

“…Humans can be involved in the process to achieve better results [16]. Many approaches make use of content descriptions associated with the table (e.g., information in an HTML page [6,17,18], headers within tables [7,17]) or rely on data in textual columns within the table to assign semantic labels [10]. Others match table rows to KB entities, leaving out of the scope matching the table columns to KB properties [19,20].…”

Section: Related Workmentioning

confidence: 99%

“…They propose learning a classifier for subject columns with features that are dependent on the name and type of the column and the values in different cells of the column. Wang et al [7] and Ermilov et al [10] proposed similar characteristics of the column: (1) the connectivity of a column (i.e., how it is connected with other columns of the table by means of properties mapped to the KB) and (2) support of the column (i.e., ratio of cells disambiguated to KB entities in the column). A combination of both connectivity and support is then used to determine which of textual column is the most likely the subject column.…”

Section: Related Workmentioning

confidence: 99%

“…This is also reflected in the benchmarks available for the problem. For instance, one of the most commonly used benchmarks, T2D [9,10], contains only 12 of 1748 tables with numerical columns disambiguated to DBpedia properties. Previous efforts [11,12] compare distributions of numerical values in columns with the distribution of literal values in a KB, matching (within a given certainty) columns to the numerical properties with the most similar distribution.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Making Sense of Numerical Data - Semantic Labelling of Web Tables

Kacprzak

Giménez-García

Piscopo

et al. 2018

Lecture Notes in Computer Science

View full text Add to dashboard Cite

With the increasing amount of structured data on the web the need to understand and support search over this emerging data space is growing. Adding semantics to structured data can help address existing challenges in data discovery, as it facilitates understanding the values in their context. While there are approaches on how to lift structured data to semantic web formats to enrich it and facilitate discovery, most work to date focuses on textual fields rather than numerical data. In this paper, we propose a two level (row and column based) approach to add semantic meaning to numerical values in tables, called NUMER. We evaluate our approach using a benchmark (NumDB) generated for the purpose of this work. We show the influence of the different levels of analysis on the success of assigning semantic labels to numerical values in tables. Our approach outperforms the state of the art and is less affected by data structure and quality issues such as a small number of entities or deviations in the data.

show abstract

Table understanding approaches for extracting knowledge from heterogeneous tables

Bonfitto

Casiraghi

Mesiti

2021

WIREs Data Min & Knowl

View full text Add to dashboard Cite

Table understanding methods extract, transform, and interpret the information contained in tabular data embedded in documents/files of different formats. Such automatic understanding would allow to exploit tabular information with the aim of accurately answering queries, or integrating heterogeneous repositories of information in a common knowledge base, or exchanging information among different sources. The purpose of this survey is to provide a comprehensive analysis of the research efforts so far devoted to the problem of table understanding and to describe systems that support the transformation of heterogeneous tables into meaningful information. This article is categorized under: Application Areas > Data Mining Software Tools Technologies > Data Preprocessing Technologies > Structure Discovery and Clustering

show abstract

TAIPAN: Automatic Property Mapping for Tabular Data

Cited by 19 publications

References 21 publications

Extracting Novel Facts from Tables for Knowledge Graph Completion

Extracting Novel Facts from Tables for Knowledge Graph Completion

Making Sense of Numerical Data - Semantic Labelling of Web Tables

Table understanding approaches for extracting knowledge from heterogeneous tables

Contact Info

Product

Resources

About