eTuner: tuning schema matching software using synthetic scenarios

Lee, Yoonkyong; Sayyadian, Mayssam; Doan, AnHai; Rosenthal, Arnon

doi:10.1007/s00778-006-0024-z

Cited by 113 publications

(98 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The learning-based approaches for system configuration fall under two types: (1) learning to classify correspondences as correct/incorrect [8,15,23]; (2) learning optimal parameter values for a system [13,16,18]. As compared to these approaches, our contribution is that we learn to select the best configuration among a set of available configurations for each matching task.…”

Section: Related Workmentioning

confidence: 99%

“…We now describe in more detail three of the machine-learning approaches [8,13,16]. The first approach learns the combination of several matchers, considered as black boxes [8].…”

Section: Related Workmentioning

confidence: 99%

“…To the best of our knowledge, the minimization of the training set size has not been investigated by others. The second approach features a technique to automatically configure a schema matching system, which has been implemented in the eTuner application [16]. This approach is interesting because it is based on the use of synthetic data obtained by perturbing some features of the schemas to be aligned.…”

Section: Related Workmentioning

confidence: 99%

“…Automatic ontology matching techniques [10] are increasingly supported by more complex systems, which use a strategy of combining several matching algorithms or matchers, each taking into account one or more ontology features. Methods that combine a set of matchers range from linear combination functions [17] to matcher cascades [7,10,25], and to arbitrary combination strategies modeled as processes where specific matchers play the role of combiners [4,16]. The choice of parameters for each matcher and of the ways in which the matchers can be combined may yield a large set of possible configurations.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Automatic Configuration Selection Using Ontology Matching Task Profiling

Cruz

Fabiani

Caimi

et al. 2012

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. An ontology matching system can usually be run with different configurations that optimize the system's effectiveness, namely precision, recall, or F-measure, depending on the specific ontologies to be aligned. Changing the configuration has potentially high impact on the obtained results. We apply matching task profiling metrics to automatically optimize the system's configuration depending on the characteristics of the ontologies to be matched. Using machine learning techniques, we can automatically determine the optimal configuration in most cases. Even using a small training set, our system determines the best configuration in 94% of the cases. Our approach is evaluated using the AgreementMaker ontology matching system, which is extensible and configurable.

show abstract

Section: Related Workmentioning

confidence: 99%

“…We now describe in more detail three of the machine-learning approaches [8,13,16]. The first approach learns the combination of several matchers, considered as black boxes [8].…”

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Automatic Configuration Selection Using Ontology Matching Task Profiling

Cruz

Fabiani

Caimi

et al. 2012

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

“…-There is a need for evaluation methods grounded on a deep analysis of the matching problem space in order to offer semi-automatic test generation methods of desired test hardness by addressing a particular point of this space (initial steps towards this line have already been done in [39]). -Despite efforts on meta-matching systems, composing matchers [18,48,50] and on Alignment API [19], ontology matching largely lacks interoperability benchmarks between tools.…”

Section: Large-scale Evaluationmentioning

confidence: 99%

Ten Challenges for Ontology Matching

Shvaiko¹,

Euzenat

2008

Lecture Notes in Computer Science

211

135

View full text Add to dashboard Cite

Abstract. This paper aims at analyzing the key trends and challenges of the ontology matching field. The main motivation behind this work is the fact that despite many component matching solutions that have been developed so far, there is no integrated solution that is a clear success, which is robust enough to be the basis for future development, and which is usable by non expert users. In this paper we first provide the basics of ontology matching with the help of examples. Then, we present general trends of the field and discuss ten challenges for ontology matching, thereby aiming to direct research into the critical path and to facilitate progress of the field.

show abstract