Record linkage: similarity measures and algorithms

By Nick Koudas, Sunita Sarawagi, Divesh Srivastava. 2006

351 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Networks and Communications, Management Science and Operations Research, Information Extraction

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 Computer Networks and Communications

In Management Science and Operations Research

In Information Extraction

Citations in the corpus

HoloClean: Holistic Data Repairs with Probabilistic Inference (2017, methodology): “For error detection, many methods rely on violations of integrity constraints [11, 14] or duplicate [27, 37, 40] and outlier detection [19, 29] methods to identify errors.”
Entity Resolution and Federated Learning get a Federated Resolution (2018, methodology): “We measure the similarity of observed shared vectors using the cosine similarity, which is a convenient similarity measure for token-based entity-resolution approaches (the other leading approaches are called edit based [19]).”
Author Name Disambiguation in MEDLINE (2009, background): “… or other possible speed-ups suggested for related computational problems such as adaptive blocking [Bilenko et al. 2006], sparse quadratic programming [Dominguez and Gonzalez-Lima 2006], approximate joins in databases [Koudas et al. 2006], and incremental (or online) clustering [Fisher 1987].”
Efficient Merging and Filtering Algorithms for Approximate String Searches (2008, methodology): “This filter is a multi-signature filter, since it produces a set of positional grams as signatures for a string.3 Prefix Filtering [10] : Given two q-gram sets G(si) and G(s2) for strings s1 and S2, we can fix an ordering 0 of the universe from which all set elements are drawn.”
Frameworks for entity matching: A comparison (2010, background): “For overview surveys and two tutorials see [60,28,3,25,35,31], respectively.”
Mining Document Collections to Facilitate Accurate Approximate Entity Matching (2009, background): “Entity extraction is important for several applications and hence received signiﬁcant attention in the literature [13, 1, 18, 16].”
Learning Semantic String Transformations from Examples (2012, background): “Most of the prior work in this area [5, 12] has focused on designing appropriate similarity functions such as edit distance, jaccard similarity, cosine similarity, and HMM25.”
Effective Data Cleaning with Continuous Evaluation (2016, background): “A large body of work has addressed some of the well-deﬁned problems; for example, refer to the survey [11] on cleaning relational data with violations of integrity constraints, and to the multiple surveys on record linkage and de-duplication [14, 9, 8, 15, 10], and also on data proﬁling [1].”
Asymmetric Minwise Hashing for Indexing Binary Inner Products and Set Containment (2015, background): “Later, because of its locality sensitive property [22], minhash became a widely used hash function for creating hash buckets leading to efficient algorithms for numerous applications including spam detection [6], collaborative filtering [4], news personalization [15], compressing social networks [13”
On indexing error-tolerant set containment (2010, methodology): “Accordingly, speciﬁc indexing methods have been proposed for the above similarity functions [14, 20, 22].”
Big data integration (2013, background): “Record linkage refers to the task of identifying records that refer to the same logical entity across different data sources, especially when they may or may not share a common identifier across the data sources [24], [19], [30], [46], [14].”
Privacy preserving schema and data matching (2007, neutral): “The deflnition of comparison function can be detailed by deflning the notion of distance which is used; examples of proposed distances include edit distance, Smith-Waterman distance and Jaro distance (see [ 22 ] for a survey).”
Asymmetric Minwise Hashing (2014, background): “…for creating hash buckets leading to efﬁcient algorithms for numerous applications including spam detection [6], collaborative ﬁltering [4], news personalization [15], compressing social networks [13], graph sampling [14], record linkage [24], duplicate detection [20], all pair similarity [5], etc.”
Evaluation of entity resolution approaches on real-world match problems (2010, neutral): “Recent surveys include [2], [20], and [24].”
Min-Max Kernels (2015, methodology): “In contrast, the resemblance kernel (2) has been widely used in practice on binary (or binarized) data [4, 5, 28, 9, 26, 7, 6, 11, 8, 16, 13, 1].”
NLyze: interactive programming by natural language for spreadsheet data analysis and manipulation (2014, background): “It would also be interesting to consider incorporating similarity matching techniques [3, 17] for column names and spread-sheet values and extending the lookup capability of NLyze to search over collections of web tables in addition to a user’s own tables (as in [35]).”
Trends in Cleaning Relational Data: Consistency and Deduplication (2015, background): “The topic has been extensively covered in many surveys [81, 40, 39, 90, 58, 70]: some aim at providing an extensive overview of all the steps involved in data deduplication [40, 58, 70], some focus on the design of similarity metrics [90, 81], some discuss the efficiency aspect of data deduplication”