Meta-mappings for schema mapping reuse

Abstract

The definition of mappings between heterogeneous schemas is a critical activity of any database application. Existing tools provide high level interfaces for the discovery of correspondences between elements of schemas, but schema mappings need to be manually specified every time from scratch, even if the scenario at hand is similar to one that has already been addressed. The problem is that schema mappings are precisely defined over a pair of schemas and cannot directly be reused on different scenarios. We tackle this challenge by generalizing schema mappings as meta-mappings: formalisms that describe transformations between generic data structures called meta-schemas. We formally characterize schema mapping reuse and explain how meta-mappings are able to: (i) capture enterprise knowledge from previously defined schema mappings and (ii) use this knowledge to suggest new mappings. We develop techniques to infer metamappings from existing mappings, to organize them into a searchable repository, and to leverage the repository to propose to users mappings suitable for their needs. We study effectiveness and efficiency in an extensive evaluation over real-world scenarios and show that our system can infer, store, and search millions of meta-mappings in seconds.

Data Vault Mappings to Dimensional Model Using Schema Matching

“…The data flow based matcher used this information at an entity level. There are possibilities to extract this data flow mapping information at an attribute level, one option is to use a tool like Queryscope 2 . This would open possibility to create more fine tuned result of the combined matcher described in this paper.…”

“…In a hybrid matcher Villányi combine a vocabulary matcher and a structural matcher, where structural matcher uses a neighborhood level structural similarity. [2]. This enables mapping reuse, when similar information is located in several schemas, whereas our reuse is use data flow mapping for creating new transformation between schemas where transformation is not yet defined.…”

“…1. Transformation stack evolutionPaolo et al introduce meta-mappings as a formalism that describes transformations between generic data structures[2]. This enables mapping reuse, when similar information is located in several schemas, whereas our reuse is use data flow mapping for creating new transformation between schemas where transformation is not yet defined.Golfarelli et al[5] introduce a starry vault approach to generate a dimensional model automatically from a data vault model.…”

SMAT: An Attention-Based Deep Learning Solution to the Automation of Schema Matching

“…Similarity can be measured by data types and domains, key characteristics (e.g., unique, primary, foreign), and relationship cardinality [1,14,29]. Recently, [3] proposes a hybrid of the constraint-based approach using key characteristics and the instance itself to create the meta-schema. Unfortunately, such approaches cannot readily handle the n:1 scenario that can be found in schema matching.…”

“…Results for ablation experiments on Precision (P), Recall (R), and F1 (F). 17.7 50.0 26.2 26.9 30 3. 28.5 42.1 98.2 58.9 10.7 38.2 16.7 OAEI 87.8 99.9 93.5 83.0 99.9 90.7 83.8 99.9 91.2 85.9 99.9 92.4 35.9 72.5 48.0 Web-forms 79.1 99.3 88.1 75.7 96.7 84.9 76.4 93.5 84.1 70.0 99.8 82.3 32.5 68.4 44.1…”