Deep Learning for Entity Matching: A Design Space Exploration

By Sidharth Mudgal, Han Li, Theodoros Rekatsinas, AnHai Doan, Youngchoon Park, Ganesh Krishnan, Rohit Deep, Esteban Arcaute, Vijay Raghavendra. 2018

675 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Management Science and Operations Research

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Management Science and Operations Research

Citations in the corpus

AlphaClean: Automatic Generation of Data Cleaning Pipelines (2019, methodology): “In response, the research community has developed a number of sophisticated data cleaning libraries for detecting and repairing errors in large datasets [12, 16, 24, 31, 37, 41].”
A Comprehensive Survey on Automatic Knowledge Graph Construction (2023, neutral): “DeepMatcher [295] Fig.”
ALEX: An Updatable Adaptive Learned Index (2019, methodology): “Also, machine learning is used to predict arrival rate of queries [23], do entity matching [26], and to predict memory access patterns for prefetching [13].”
Toward a System Building Agenda for Data Integration (and Data Science) (2018, methodology): “Integrating Research, Education, and Outreach: Research-wise we have developed a solution for debugging the blocking step of EM [24] and matching rules [28], using deep learning to match textual and dirty data [27], scaling up EM workflows [7], and developing a novel architecture for cloud EM servic”
Neo: A Learned Query Optimizer (2019, methodology): “In fact, word2vec-based embeddings have recently been applied to other database problems, such as semantic querying [7], entity matching [41], and data discovery [12].”
Deduplication of Scholarly Documents using Locality Sensitive Hashing and Word Embeddings (2020, background): “, 2010) and more recently, deep learning (Mudgal et al., 2018) have also been proposed in this regard.”
Deep entity matching with pre-trained language models (2020, methodology): “: This is the full version of our system with all 3 optimizations”
Distributed Representations of Tuples for Entity Resolution (2017, positive): “echniques are modular enough and can be easily incorporated into any of the existing systems. There has been extensive interest in applying DL in data cleaning [48]. A recent work that extends DeepER [41] explored the design space of ER using DRs. The authors introduce four dierent choices for the a”
A benchmarking study of embedding-based entity alignment for knowledge graphs (2020, neutral): “hes for (heterogeneous information) network alignment [29, 44, 89] or cross-lingual knowledge projection [56], which may also be modied for entity alignment. It is also worth noting that two studies [14, 52] design the embeddingbased approaches for entity resolution in databases. They represent the ”
Aligning Cross-Lingual Entities with Multi-Aspect Information (2019, methodology): “The reason we use highway networks is consistent with the conclusions of Mudgal et al. (2018), who conducted a design space exploration of neural models for entity matching and found that highway networks are generally better than FC layers in convergence speed and effectiveness.”
Neural Databases (2020, background): “Based on these successes, researchers have considered the application of neural nets to data management problems, including learning indices [21], query optimization and entity matching [25, 29].”
Can Foundation Models Wrangle Your Data? (2022, background): “EM has been extensively studied over the past decade (see [64] for a survey) and methods broadly fall into three categories: rulebased, crowd-based [31, 81] and ML/DL-based [41, 62].”
A Survey of Deep Active Learning (2020, background): “[34, 109, 197, 199] CoNLL [184], NCBI [51], MedMentions [149], OntoNotes [167], DBLP, FZ, AG [147], Cora [228] Named entity recognition (NER) Question answering [13] CMDC [43], JabberWacky’s chatlogs9 Dialogue generation [104] Visual Genome [117], VQA [11] Visual question answer (VQA)”