Neural Networks for Entity Matching: A Survey

Barlaug, Nils; Gulla, Jon Atle

doi:10.1145/3442200

Cited by 68 publications

(35 citation statements)

References 81 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Both propose a solution to challenge (2), but as we discussed above, with important limitations. Neither addresses challenge (3). Moreover, they do not provide a unified and coherent way of actually communicating or visualizing an explanation to the end-user in the same way the original authors of LIME do -something we aim to do.…”

Section: Related Workmentioning

confidence: 96%

“…While early work focused on custom architectures and trained models from scratch, the current state of the art focuses on fine-tuning large natural language models such as BERT [7], which offers higher accuracy and decreases the need for training examples [18]. We refer to Barlaug and Gulla [3] for an extensive survey on deep learning for EM.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

LEMON: Explainable Entity Matching

Barlaug¹

2021

Preprint

Self Cite

View full text Add to dashboard Cite

State-of-the-art entity matching (EM) methods are hard to interpret, and there is significant value in bringing explainable AI to EM. Unfortunately, most popular explainability methods do not work well out of the box for EM and need adaptation. In this paper, we identify three challenges of applying local post hoc feature attribution methods to entity matching: cross-record interaction effects, nonmatch explanations, and variation in sensitivity. We propose our novel model-agnostic and schema-flexible method LEMON 1 that addresses all three challenges by (i) producing dual explanations to avoid cross-record interaction effects, (ii) introducing the novel concept of attribution potential to explain how two records could have matched, and (iii) automatically choosing explanation granularity to match the sensitivity of the matcher and record pair in question. Experiments on public datasets demonstrate that the proposed method is more faithful to the matcher and does a better job of helping users understand the decision boundary of the matcher than previous work. Furthermore, user studies show that the rate at which human subjects can construct counterfactual examples after seeing an explanation from our proposed method increases from 54% to 64% for matches and from 15% to 49% for non-matches compared to explanations from a standard adaptation of LIME.

show abstract

Section: Related Workmentioning

confidence: 96%

Section: Related Workmentioning

confidence: 99%

LEMON: Explainable Entity Matching

Barlaug¹

2021

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…In MinHash LSH, we configured the number of bands and rows, restricting both of them to powers of two such that their product is also a power of two, i.e., 2 𝑛 with 𝑛 ∈ {7, 8, 9}. For 𝑘-shingles, we considered four common values for 𝑘, i.e., [2,5].…”

Section: Qualitative Analysismentioning

confidence: 99%

“…All experiments were run on Java 15. For MH-LSH, we used java-LSH, version 0.12 5 . For HP-and CP-LSH, we used the Python wrapper of FALCONN [1], version 1.3.1 6 .…”

Section: Quantitative Analysismentioning

confidence: 99%

“…Early attempts were rule-based, comparing similarity values with thresholds, but more recent techniques rely on learning, i.e., they usually model matching as a binary classification task (match, non-match) [51]. Using a labelled training dataset, (semi-)supervised, active, and deep learning techniques are adapted to ER [5]. In some cases, a clustering step is subsequently applied on the resulting similarity scores in order to refine the output [29].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

How to reduce the search space of Entity Resolution: with Blocking or Nearest Neighbor search?

Papadakis

Fisichella

Schoger

et al. 2022

Preprint

View full text Add to dashboard Cite

Entity Resolution suffers from quadratic time complexity. To increase its time efficiency, three kinds of filtering techniques are typically used for restricting its search space: (i) blocking workflows, which group together entity profiles with identical or similar signatures, (ii) string similarity join algorithms, which quickly detect entities more similar than a threshold, and (iii) nearest-neighbor methods, which convert every entity profile into a vector and quickly detect the closest entities according to the specified distance function. Numerous methods have been proposed for each type, but the literature lacks a comparative analysis of their relative performance. As we show in this work, this is a non-trivial task, due to the significant impact of configuration parameters on the performance of each filtering technique. We perform the first systematic experimental study that investigates the relative performance of the main methods per type over 10 real-world datasets. For each method, we consider a plethora of parameter configurations, optimizing it with respect to recall and precision. For each dataset, we consider both schema-agnostic and schema-based settings. The experimental results provide novel insights into the effectiveness and time efficiency of the considered techniques, demonstrating the superiority of blocking workflows and string similarity joins.

show abstract

Deep Learning Applications for Interactive Marketing in the Contemporary Digital Age

2023

The Palgrave Handbook of Interactive Marketing

View full text Add to dashboard Cite

Neural Networks for Entity Matching: A Survey

Cited by 68 publications

References 81 publications

LEMON: Explainable Entity Matching

LEMON: Explainable Entity Matching

How to reduce the search space of Entity Resolution: with Blocking or Nearest Neighbor search?

Deep Learning Applications for Interactive Marketing in the Contemporary Digital Age

Contact Info

Product

Resources

About