Learning object identification rules for information integration

Tejada, Sheila; Knoblock, Craig A.; Minton, Steven

doi:10.1016/s0306-4379(01)00042-4

Cited by 231 publications

(149 citation statements)

References 35 publications

Supporting

Mentioning

148

Contrasting

Unclassified

Order By: Relevance

“…These string metrics have been developed and applied in different scientific fields like statistics, for probabilistic record linkage [6], database, for record matching [7], Artificial Intelligence, for supervised learning [8], and Biology, for identifying common molecular subsequences [9]. In the current paper we have considered the Levenstein [10] distance, which counts the insertions and deletions needed to match two strings, the Needleman-Wunsch [11] distance, which assigns a different cost on the edit operations, the Smith-Waterman [9], which additionally uses an alphabet mapping to costs and the Monge-Elkan [7], which uses variable costs depending on the substring gaps between the words.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

A String Metric for Ontology Alignment

Stoilos

Stamou

Kollias

2005

Lecture Notes in Computer Science

301

165

View full text Add to dashboard Cite

Abstract. Ontologies are today a key part of every knowledge based system. They provide a source of shared and precisely defined terms, resulting in system interoperability by knowledge sharing and reuse. Unfortunately, the variety of ways that a domain can be conceptualized results in the creation of different ontologies with contradicting or overlapping parts. For this reason ontologies need to be brought into mutual agreement (aligned). One important method for ontology alignment is the comparison of class and property names of ontologies using stringdistance metrics. Today quite a lot of such metrics exist in literature. But all of them have been initially developed for different applications and fields, resulting in poor performance when applied in this new domain. In the current paper we present a new string metric for the comparison of names which performs better on the process of ontology alignment as well as to many other field matching problems.

show abstract

Section: Related Workmentioning

confidence: 99%

“…Thus we could not resist but to evaluate it with classical benchmarks found in literature like the ones in [7,8,24,20]. The list of the datasets used can be found in Table 2 as well as the number of strings that each dataset includes.…”

Section: Census and Field Matchingmentioning

confidence: 99%

A String Metric for Ontology Alignment

Stoilos

Stamou

Kollias

2005

Lecture Notes in Computer Science

301

165

View full text Add to dashboard Cite

show abstract

“…BN [38] MOMA [55] SERF [5] Active Atlas [53,54] MARLIN [11,12] Multiple Classifier System [62] Operator Trees [13] TAILOR [24] FEBRL [18,17] STEM [36] Context Based Framework [16] Training-based between two entities. The previously proposed approaches mostly assume that corresponding attributes from the input datasets have been determined beforehand, either manually or with the help of schema matching.…”

Section: Matchersmentioning

confidence: 99%

“…The use of supervised (training-based) approaches or learners aims at automating the process of entity matching to reduce the required manual effort. Training-based approaches, e.g., Naïve Bayes [49], logistic regression [46], Support Vector Machine (SVM) [11,43,49] or decision trees [63,29,49,53,54,56] have so far been used for some subtasks, e.g., determining suitable parameterizations for matchers or adjusting combination functions parameters (weights for matchers, offsets). However, training-based approaches require suitable training data and providing such data typically involves manual effort.…”

Section: Combination Of Matchersmentioning

confidence: 99%

“…Three of the 11 frameworks operate without training; four of the eight training-based frameworks, the Context Based Framework [16], FEBRL [18,17], STEM [36] and TAILOR [24], also support the manual specification of EM strategies without training (hybrid frameworks). Four of the training-based frameworks expect the users to provide suitable training data manually, while Active Atlas [53,54], MARLIN [11,12], STEM [36] and FEBRL [18,17] have also investigated (semi-)automatic training methods.…”

Section: Training Selectionmentioning

confidence: 99%

See 1 more Smart Citation

Frameworks for entity matching: A comparison

Köpcke

Rahm

2010

Data & Knowledge Engineering

346

204

View full text Add to dashboard Cite

a b s t r a c tEntity matching is a crucial and difficult task for data integration. Entity matching frameworks provide several methods and their combination to effectively solve different match tasks. In this paper, we comparatively analyze 11 proposed frameworks for entity matching. Our study considers both frameworks which do or do not utilize training data to semiautomatically find an entity matching strategy to solve a given match task. Moreover, we consider support for blocking and the combination of different match algorithms. We further study how the different frameworks have been evaluated. The study aims at exploring the current state of the art in research prototypes of entity matching frameworks and their evaluations. The proposed criteria should be helpful to identify promising framework approaches and enable categorizing and comparatively assessing additional entity matching frameworks and their evaluations.

show abstract

A generic Web-based entity resolution framework

Pereira

Ribeiro-Neto

Ziviani

et al. 2011

J. Am. Soc. Inf. Sci.

View full text Add to dashboard Cite

Web data repositories usually contain references to thousands of real-world entities from multiple sources. It is not uncommon that multiple entities share the same label (polysemes) and that distinct label variations are associated with the same entity (synonyms), which frequently leads to ambiguous interpretations. Further, spelling variants, acronyms, abbreviated forms, and misspellings compound to worsen the problem. Solving this problem requires identifying which labels correspond to the same real-world entity, a process known as entity resolution. One approach to solve the entity resolution problem is to associate an authority identifier and a list of variant forms with each entity-a data structure known as an authority file. In this work, we propose a generic framework for implementing a method for generating authority files. Our method uses information from the Web to improve the quality of the authority file and, because of that, is referred to as WER-Web-based Entity Resolution. Our contribution here is threefold: (a) we discuss how to implement the WER framework, which is flexible and easy to adapt to new domains; (b) we run extended experimentation with our WER framework to show that it outperforms selected baselines; and (c) we compare the results of a specialized solution for author name resolution with those produced by the generic WER framework, and show that the WER results remain competitive.1. for each record do: (a) select one or more attributes and use them to compose a query, (b) submit the query to a search engine, and (c) collect the top m documents in the answer set; 2. parse the documents in the answer set and extract (Webinferred) attributes B j such as URLs, titles, texts of the documents, names, and acronyms used to refer to the entities; 3. using the original attributes A i and the Web-inferred attributes B j , cluster the entity records relative to the primary entity e pr , such that each cluster corresponds to a single real-world entity; 4. in each cluster, select an attribute value to be the canonical entity name, and output the respective entry to the authority file.

show abstract

Learning object identification rules for information integration

Cited by 231 publications

References 35 publications

A String Metric for Ontology Alignment

A String Metric for Ontology Alignment

Frameworks for entity matching: A comparison

A generic Web-based entity resolution framework

Contact Info

Product

Resources

About