contrastive entity linkage mining variational attributes
play

Contrastive Entity Linkage : Mining Variational Attributes from - PowerPoint PPT Presentation

Aug 15, 2023 •436 likes •603 views

Contrastive Entity Linkage : Mining Variational Attributes from Large Catalogs for Entity Linkage AKBC 2020 Varun Embar , Bunyamin Sisman, Hao Wei, Xin Luna Dong, Christos Faloutsos and Lise Getoor Motivation iPhone 11 Pro 64 GB iPhone 11 Pro

  1. Contrastive Entity Linkage : Mining Variational Attributes from Large Catalogs for Entity Linkage AKBC 2020 Varun Embar , Bunyamin Sisman, Hao Wei, Xin Luna Dong, Christos Faloutsos and Lise Getoor

  2. Motivation iPhone 11 Pro 64 GB iPhone 11 Pro 256 GB Are these two entities the same or different ?

  3. Motivation Attributes Same Brand Color Generation iPhone 11 Pro 64 GB Different iPhone 11 Pro 256 GB Storage

  4. Motivation Same Brand Base Manufacturer Attributes Model Variations iPhone 11 Pro 64 GB Different Color Variational Storage Attributes iPhone 11 Pro 128 GB

  5. Motivation Duplicates , apple 11 amazon 5 Distinct bose qcII Entity Linkage , Catalog 1 Variations bose qcII apple 11 , bose qcIII Catalog 2

  6. Contributions [C1] Automatic variational attribute discovery ○ Propose contrast feature that model variation attributes ○ Novel scalable, unsupervised VarSpot algo to extract them [C2] Three-way entity linkage ○ Distinct, variation and duplicates ○ Contrastive entity linkage framework [C3] Effectiveness ○ Empirical evaluation on three different domains ○ Three different entity linkage frameworks

  7. Related Work Duplicate Variation Variational Attribute Matching Matching Extraction Entity Linkage Approaches[1] GROUP Li et al. [2015] Recasens et al. [2011] Attribute Extraction Techniques [2] Contrastive Entity Linkage [1] Christen et. al. 2012, Rahm, 2010, Halevy 2005, Machanavajjhala 2012 etc. [2] Zheng 2018, Bizer 2017, Weld 2012, Hu 2011, Kannan 2011 etc.

  8. Approach - VarSpot C1 Phase 1 apple 11 , amazon 5 bose qcII Blocking & Catalog 1 , Linkage apple 11 Same Catalog amazon 5 , bose qcII Catalog 1 See paper for more details

  9. Approach - VarSpot C1 Phase 2 Apple iPhone 11 Pro 64 GB Apple iPhone 11 Pro 256 GB Contrast features

  10. Approach - Contrastive entity linkage C2 Duplicates apple 11 white , amazon 5 black bose qcII black Entity Distinct Catalog 1 linkage framework , bose qcII rose Variations apple 11 black , bose qcIII black Catalog 2 Extracted contrast features

  11. Evaluation C3 Domains ● Software (Small-sized dataset) ● Groceries (Medium-sized dataset) ● Music (Large-sized dataset) Entity linkage frameworks ● Magellan [Konda et. al. 2016] ● SILK [Isele et. al. 2010] ● Deepmatcher [Mudgal et. al. 2018]

  12. Evaluation C3 Variations identified by VarSpot algorithm Groceries Music Milk duds candy 1.85 ounce boxes pack of 24 Groove is in the heart Milk duds candy 5 ounce boxes pack of 3 Groove is in the heart club version Milk duds movie size 5 oz 12 count Groove is in the heart sampladelic remix Software Peachtree by sage premium accounting for nonprofits 2007 Peachtree by sage premium accounting 2007 accountants’ edition Peachtree by sage pro accounting 2007

  13. Evaluation C3 Top contrast features identified by VarSpot algorithm Software Groceries Music standard mac pack of 6 remix upgrade small box pack of 2 mix premium upsell mac 2 pack radio edit standard upsell mac red live deluxe strawberry instrumental

  14. Evaluation C3 Magellan Software Without contrast CEL features Duplicates F1 0.785 0.81 APS 0.877 0.897 Variations F1 0.677 0.695 APS 0.761 0.777 CEL significantly outperform models without contrast features More results in the paper

  15. For more details visit our poster # fR44nF03Rb

Download Presentation
Download Policy: The content available on the website is offered to you 'AS IS' for your personal information and use only. It cannot be commercialized, licensed, or distributed on other websites without prior consent from the author. To download a presentation, simply click this link. If you encounter any difficulties during the download process, it's possible that the publisher has removed the file from their server.

Recommend

From E/R Diagrams to Relations Entity set relation Attributes attributes

From E/R Diagrams to Relations Entity set relation Attributes attributes

From E/R Diagrams to Relations Entity set relation Attributes attributes Relationships relations whose attributes are only: The keys of the connected entity sets Attributes of the relationship itself 1 Entity Set

846 views • 70 slides
Entity-Relationship Models 1 / 24 Entity-Relationship Models Entities Attributes

Entity-Relationship Models 1 / 24 Entity-Relationship Models Entities Attributes

Entity-Relationship Models 1 / 24 Entity-Relationship Models Entities Attributes Relationships 2 / 24 The Role of Conceptual Models High-level but concrete view of data understandable by end users and database developers 3 / 24

628 views • 24 slides
This Lecture Entity/Relationship models Entities and Attributes Entity Relationship

This Lecture Entity/Relationship models Entities and Attributes Entity Relationship

This Lecture Entity/Relationship models Entities and Attributes Entity Relationship Modelling Relationships E/R Diagrams Further Reading Database Systems Database Systems, Connolly & Begg, Chapter 12 Michael Pound

172 views • 6 slides
Entity Linkage for Heterogeneous, Uncertain, and Volatile Data Ekaterini Ioannou L3S Research

Entity Linkage for Heterogeneous, Uncertain, and Volatile Data Ekaterini Ioannou L3S Research

Introduction LinkDB Query Processing Detecting Linkages Conclusions Entity Linkage for Heterogeneous, Uncertain, and Volatile Data Ekaterini Ioannou L3S Research Center Leibniz Universit at Hannover Friday, 15th of April, 2011

1.14k views • 78 slides
Probabilistic Data Generation for Deduplication and Data Linkage Peter Christen Data Mining

Probabilistic Data Generation for Deduplication and Data Linkage Peter Christen Data Mining

Probabilistic Data Generation for Deduplication and Data Linkage Peter Christen Data Mining Group, Australian National University Contact: peter.christen@anu.edu.au Project web page: http://datamining.anu.edu.au/linkage.html Funded by the ANU,

238 views • 13 slides
Contrastive Causation Making Causation Contrastive What this talk presupposes... The

Contrastive Causation Making Causation Contrastive What this talk presupposes... The

Contrastive Causation Making Causation Contrastive What this talk presupposes... The counterfactual account of causation ... and its discontents (counterexamples) Recently, some philosophers have tried to solve

625 views • 33 slides
The language of visual attributes Kristen Grauman Facebook AI Research University of Texas at

The language of visual attributes Kristen Grauman Facebook AI Research University of Texas at

The language of visual attributes Kristen Grauman Facebook AI Research University of Texas at Austin Attributes vs. objects Red Round Visual Physical properties entity Ripe Fresh Value of attributes Find a more Zebras have stripes

664 views • 51 slides
Multi-Source Spatial Entity Linkage Suela Isaj Supervisor: Torben Bach Pedersen (AAU)

Multi-Source Spatial Entity Linkage Suela Isaj Supervisor: Torben Bach Pedersen (AAU)

Multi-Source Spatial Entity Linkage Suela Isaj Supervisor: Torben Bach Pedersen (AAU) Co-supervisor: Esteban Zimnyi (ULB) 1 Multi-Source Spatial Entities 2 Overall PhD study 3 Geo-social related work Old datasets Non-operational

651 views • 29 slides
What is data (or record) linkage? Recent interest in data linkage The process of linking and

What is data (or record) linkage? Recent interest in data linkage The process of linking and

Data Linkage Techniques: What is data linkage? Past, Present and Future Applications and challenges Peter Christen The past A short history of data linkage Department of Computer Science, The Australian National University The present

428 views • 4 slides
Transformations for Record Linkage Matthew Michelson & Craig A. Knoblock Fetch

Transformations for Record Linkage Matthew Michelson & Craig A. Knoblock Fetch

Mining the Heterogeneous Transformations for Record Linkage Matthew Michelson & Craig A. Knoblock Fetch Technologies USC Information Sciences Institute ICAI 2009 Record Linkage Source 1 Manager Restaurant Bobby Jones California

436 views • 15 slides
Introduction to Data Science: Data K K K |Batting| |Master| + | B 1| | M 1| + 2| R | | T

Introduction to Data Science: Data K K K |Batting| |Master| + | B 1| | M 1| + 2| R | | T

Overview Formal introduction to keys Entity Resolution and Record Linkage Database Query Optimization The Entity-Relationship and Relational Models Data representational modeling Database Query Optimization Entity Resolution and Record

1.7k views • 91 slides
Probabilistic Deduplication, Data Linkage and Geocoding Peter Christen Data Mining Group,

Probabilistic Deduplication, Data Linkage and Geocoding Peter Christen Data Mining Group,

Probabilistic Deduplication, Data Linkage and Geocoding Peter Christen Data Mining Group, Australian National University in collaboration with Centre for Epidemiology and Research, New South Wales Department of Health Contact:

629 views • 44 slides
Automating variational inference for statistics and data mining Tom Minka Machine Learning and

Automating variational inference for statistics and data mining Tom Minka Machine Learning and

Automating variational inference for statistics and data mining Tom Minka Machine Learning and Perception Group Microsoft Research Cambridge A common situation You have a dataset Some models in mind Want to fit many different

607 views • 31 slides
Enhanced Entity-Relationship Models 1 / 10 Enahnced Entity-Relationship Models Enhanced ER

Enhanced Entity-Relationship Models 1 / 10 Enahnced Entity-Relationship Models Enhanced ER

Enhanced Entity-Relationship Models 1 / 10 Enahnced Entity-Relationship Models Enhanced ER models add inheritance to entity types superclasses and subclasses. Subclasses inherit all the attributes of their superclass(es). Subclasses can:

420 views • 10 slides
C ONCEPTUAL D ESIGN : ER TO R ELATIONAL TO SQL How to represent Entity sets,

C ONCEPTUAL D ESIGN : ER TO R ELATIONAL TO SQL How to represent Entity sets,

R ELATIONAL M ODEL TO SQL Data Model C ONCEPTUAL D ESIGN : ER TO R ELATIONAL TO SQL How to represent Entity sets, Relationship sets, Attributes, Key and participation constraints, Subclasses, Weak entity sets . . . ?

431 views • 12 slides
Febrl A parallel open source record linkage and geocoding system Peter Christen Data Mining

Febrl A parallel open source record linkage and geocoding system Peter Christen Data Mining

Febrl A parallel open source record linkage and geocoding system Peter Christen Data Mining Group, Australian National University in collaboration with Centre for Epidemiology and Research, New South Wales Department of Health Contact:

404 views • 36 slides
Moscow, Russia EFFECTIVE INVESTMENTS IN RUSSIAS GOLD MINING INDUSTRY 2013 Lou Naumovski, VP

Moscow, Russia EFFECTIVE INVESTMENTS IN RUSSIAS GOLD MINING INDUSTRY 2013 Lou Naumovski, VP

Kinross Gold Corporation: February 5 Moscow, Russia EFFECTIVE INVESTMENTS IN RUSSIAS GOLD MINING INDUSTRY 2013 Lou Naumovski, VP and General Director, Moscow Representative Office Kinross Gold Corporation 1 1 February 5, 2013

363 views • 22 slides
Introduction to Graphs Russell Impagliazzo and Miles Jones Thanks to Janine Tiefenbruck

Introduction to Graphs Russell Impagliazzo and Miles Jones Thanks to Janine Tiefenbruck

Introduction to Graphs Russell Impagliazzo and Miles Jones Thanks to Janine Tiefenbruck http://cseweb.ucsd.edu/classes/sp16/cse21-bd/ April 18, 2016 1 What is a graph? A (directed) graph G is A nonempty set of vertices V, also called

755 views • 40 slides
Solving Systems of Equa3ons Using Mul3ple Strategies NCTM Interac3ve Ins3tute, 2016 Name

Solving Systems of Equa3ons Using Mul3ple Strategies NCTM Interac3ve Ins3tute, 2016 Name

Solving Systems of Equa3ons Using Mul3ple Strategies NCTM Interac3ve Ins3tute, 2016 Name Title/Posi3on Affilia3on Email Address Common Core Standards This session will address the following: 8.EE.8a Understand that solu8ons to a system of

482 views • 24 slides
Q1 2020 Earnings Review May 7, 2020 1 Cautionary ry Note Non-GAAP Measures This presentation

Q1 2020 Earnings Review May 7, 2020 1 Cautionary ry Note Non-GAAP Measures This presentation

Q1 2020 Earnings Review May 7, 2020 1 Cautionary ry Note Non-GAAP Measures This presentation of Pan American Silver Corp. and its subsidiaries (collectively, Pan American, Pan American Silver, the Company, we or

691 views • 24 slides
Announcements HW 2 is due next Tuesday No class on next Tuesday, but TAs will be here to

Announcements HW 2 is due next Tuesday No class on next Tuesday, but TAs will be here to

Announcements HW 2 is due next Tuesday No class on next Tuesday, but TAs will be here to collect HW 1 CS6501: T opics in Learning and Game Theory (Fall 2019) Prediction Markets (as a Forecasting T ool) Instructor: Haifeng Xu Slides of

884 views • 46 slides
Theory and Algorithms of Physical Design C.K. Cheng, T.C. Hu, and A.B. Kahng CSE department UC

Theory and Algorithms of Physical Design C.K. Cheng, T.C. Hu, and A.B. Kahng CSE department UC

Theory and Algorithms of Physical Design C.K. Cheng, T.C. Hu, and A.B. Kahng CSE department UC San Diego 1 O UTLINES Biography of T.C. Hu Physical Design Column Generating Techniques Motto Sequencing Picks

485 views • 24 slides
1 What is a gold standard? A monetary system where the government links its supply of paper

1 What is a gold standard? A monetary system where the government links its supply of paper

1 What is a gold standard? A monetary system where the government links its supply of paper money to a stock of gold reserves (visual) Under a GS, the government de fi nes a dollar to be a speci fi c weight of gold A dollar bill is a

579 views • 12 slides
www.FLgov.com/FBCB Special Thanks To: Ms. Dot Johnson Mr. Mark Retman Office of the Mayor

www.FLgov.com/FBCB Special Thanks To: Ms. Dot Johnson Mr. Mark Retman Office of the Mayor

www.FLgov.com/FBCB Special Thanks To: Ms. Dot Johnson Mr. Mark Retman Office of the Mayor Information Technology Mr. Tom Bronakoski WCOT Special Thanks To: www.heartgallerybigbend.org Special Thanks To: Executive Office of the Governor

677 views • 33 slides

More recommend