Sparkly: A Simple yet Surprisingly Strong TF/IDF Blocker for Entity Matching

Summary: Sparkly: a distributed, share-nothing top-k TF/IDF blocker built on Lucene/Spark with automatic attribute and tokenizer selection for entity matching. Outperforms eight state-of-the-art blockers on recall/output-size/runtime, advocating TF/IDF and top-k blocking as strong, scalable baselines. (summarized by gpt-5-mini on Feb 09 2026)

Paper ID: hacbc6e5a8a80d96c
Venue: VLDB
Year: 2023
Pagerank: 5.8189161e-05
Overall Rank: 6,300 | 57.65%
DOI: 10.14778/3583140.3583163

Incoming Non-self Citations Over Time

Authors

1. Derek Paulsen (Informatica; University of Wisconsin)
2. Yash Govind (Apple)
3. AnHai Doan (Informatica; University of Wisconsin)

BibTeX Citation

@article{paulsen_vldb23,
        title = {{Sparkly: A Simple yet Surprisingly Strong TF/IDF Blocker for Entity Matching}},
        author = {Paulsen, Derek and Govind, Yash and Doan, AnHai},
        journal = {PVLDB},
        series = {{VLDB} '23},
        volume = {16},
        number = {6},
        pages = {1507--1519},
        doi = {10.14778/3583140.3583163},
        url = {https://doi.org/10.14778/3583140.3583163},
        year = {2023}
}

Incoming Citations (Sorted by Pagerank)

Showing 6 of 6 citing papers.

Rank	Citing Paper	Year	Venue	Pagerank
9,565	Deduplicated Sampling On-Demand	2025	VLDB	5.1571823e-05
10,180	Progressive Entity Matching: A Design Space Exploration	2025	SIGMOD	5.0653015e-05
10,462	Generalized Entity Matching with Adaptivity via Large Language Models	2026	SIGMOD	4.9793485e-05
10,538	3dSAGER: Geospatial Entity Resolution over 3D Objects	2026	SIGMOD	4.9793485e-05
10,745	Balancing the Blend: An Experimental Analysis of Trade-offs in Hybrid Search	2026	VLDB	4.9793485e-05
10,748	ALER: An Active Learning Hybrid System for Efficient Entity Resolution	2026	VLDB	4.9793485e-05

Outgoing Citations (Sorted by Pagerank)

Showing 12 of 12 cited papers.

Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.

Rank	Cited Paper	Year	Venue	Pagerank
134	Deep Entity Matching with Pre-Trained Language Models	2021	VLDB	0.00030043481
158	Deep Learning for Entity Matching: A Design Space Exploration	2018	SIGMOD	0.00028046388
161	Robust and Efficient Fuzzy Match for Online Data Cleaning	2003	SIGMOD	0.00027718195
433	Corleone: Hands-Off Crowdsourcing for Entity Matching	2014	SIGMOD	0.00018332741
457	Distributed Representations of Tuples for Entity Resolution	2018	VLDB	0.00017907103
530	Magellan: Toward Building Entity Matching Management Systems	2016	VLDB	0.00016855162
1,654	Falcon: Scaling Up Hands-Off Crowdsourced Entity Matching to Build Cloud Services	2017	SIGMOD	9.9739611e-05
2,514	Deep Learning for Blocking in Entity Matching: A Design Space Exploration	2021	VLDB	8.3648432e-05
4,337	Algorithmic Aspects of Parallel Query Processing	2018	SIGMOD	6.6541797e-05
4,970	Auto-FuzzyJoin: Auto-Program Fuzzy Similarity Joins Without Labeled Examples	2021	SIGMOD	6.3329595e-05
7,509	Entity Matching Meets Data Science: A Progress Report from the Magellan Project	2019	SIGMOD	5.5062607e-05
12,244	CloudMatcher: A Hands-Off Cloud/Crowd Service for Entity Matching	2018	VLDB	4.9793485e-05

Semantically Similar Papers

#	Overall Rank	Paper	Year	Venue
1	4,513	LocationSpark: A Distributed In-Memory Data Management System for Big Spatial Data	2016	VLDB
2	9,458	HyperBlocker: Accelerating Rule-based Blocking in Entity Resolution using GPUs	2025	VLDB
3	11,880	Generalized Supervised Meta-blocking	2022	VLDB
4	942	SPARK: Top-k Keyword Query in Relational Databases	2007	SIGMOD
5	5,143	Schema-agnostic vs Schema-based Configurations for Blocking Methods on Homogeneous Data	2016	VLDB
6	1,307	Entity Resolution with Iterative Blocking	2009	SIGMOD
7	5,064	Supervised Meta-blocking	2014	VLDB
8	2,514	Deep Learning for Blocking in Entity Matching: A Design Space Exploration	2021	VLDB
9	3,738	BLAST: a Loosely Schema-aware Meta-blocking Approach for Entity Resolution	2016	VLDB
10	2,110	Comparative Analysis of Approximate Blocking Techniques for Entity Resolution	2016	VLDB