Generating Unified Famous Objects (UFOs) from the classified object tables

Kola, Anusha; More, Harshal; Soderman, Sean; Gubanov, Michael

doi:10.1109/bigdata.2017.8258537

Cited by 4 publications

(7 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Section 4 describes a search scenario & classification scenario that illustrate our system. UFOs: Unified Famous Objects, are a structure used to abstract away differences in data representation [27,33,41]. See Section 5 for a more in-depth discussion on UFOs.…”

Section: • a Learning Search Engine For Large-scale Structuredmentioning

confidence: 99%

“…Our system focuses on generating these sets of core properties using term matching to count their occurrences [41]. We use them to assert the relevance of rows within the dataset, rather than building UFOs for identifying similar objects.…”

Section: Related Workmentioning

confidence: 99%

“…Attributes in different languages are still useful for gathering data that is associated with a certain object (here, "songs"). Refer to [27,32,33,41] on UFO definition and more details on automatic UFO construction and data fusion using UFOs.…”

Section: Unified Famous Object -Songsmentioning

confidence: 99%

“…In this scheme, we first extract UFO core attributes [33,41] from each Web tables' subset labeled by our trained classifier to belong to a certain class (e.g. songs), then use them to retrieve the most relevant search results from the entire corpus.…”

Section: Using Trained Classifiers and Unifiedmentioning

confidence: 99%

“…Informally core attributes are the most important attributes of an object (e.g. wings for a bird) a critical component of UFOs, a data structure used for fusing similar structured data objects that are represented in different ways [27,41]. We use core attributes as soft constraints, improving the rank of results that are related to a specific object of interest.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Hybrid.AI

Soderman

Kola

Podkorytov

et al. 2018

Companion of the the Web Conference 2018 on the Web Conference 2018 - WWW '18

Self Cite

View full text Add to dashboard Cite

Variety of Big data [17, 40, 44, 47, 52] is a significant impediment for anyone who wants to search inside a large-scale structured dataset. For example, there are millions of tables available on the Web, but the most relevant search result does not necessarily match the keyword-query exactly due to a variety of ways to represent the same information. Here we describe Hybrid.AI, a learning search engine for largescale structured data that uses automatically generated machine learning classifiers and Unified Famous Objects (UFOs) [33] to return the most relevant search results from a large-scale Web tables corpora. We evaluate it over this corpora, collecting 99 queries and their results from users, and observe significant relevance gain.

show abstract

Section: • a Learning Search Engine For Large-scale Structuredmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Unified Famous Object -Songsmentioning

confidence: 99%

Section: Using Trained Classifiers and Unifiedmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Hybrid.AI

Soderman

Kola

Podkorytov

et al. 2018

Companion of the the Web Conference 2018 on the Web Conference 2018 - WWW '18

Self Cite

View full text Add to dashboard Cite

show abstract

PROCLAIM: An Unsupervised Approach to Discover Domain-Specific Attribute Matchings from Heterogeneous Sources

Arman

Wlodarczyk

Seghouani

et al. 2020

Advanced Information Systems Engineering

View full text Add to dashboard Cite

Schema matching is a critical problem in many applications where the main goal is to match attributes coming from heterogeneous sources. In this paper, we propose PROCLAIM (PROfile-based Cluster-Labeling for AttrIbute Matching), an automatic, unsupervised clustering-based approach to match attributes of a large number of heterogeneous sources. We define the concept of attribute profile to characterize the main properties of an attribute using: (i) the statistical distribution and the dimension of the attribute's values, (ii) the name and textual descriptions related to the attribute. The attribute matchings produced by PROCLAIM give the best representation of heterogeneous sources thanks to the cluster-labeling function we defined. We evaluate PROCLAIM on 45,000 different data sources coming from oil and gas authority open data website3. The results we obtain are promising and validate our approach.

show abstract

Big Data Security Challenges and Preventive Solutions

Gupta¹,

Rohil

2019

Data Management, Analytics and Innovation

View full text Add to dashboard Cite

Generating Unified Famous Objects (UFOs) from the classified object tables

Cited by 4 publications

References 8 publications

Hybrid.AI

Hybrid.AI

PROCLAIM: An Unsupervised Approach to Discover Domain-Specific Attribute Matchings from Heterogeneous Sources

Big Data Security Challenges and Preventive Solutions

Contact Info

Product

Resources

About