RODI: Benchmarking relational-to-ontology mapping generation quality

Abstract

Accessing and utilizing enterprise or Web data that is scattered across multiple data sources is an important task for both applications and users. Ontology-based data integration, where an ontology mediates between the raw data and its consumers, is a promising approach to facilitate such scenarios. This approach crucially relies on useful mappings to relate the ontology and the data, the latter being typically stored in relational databases. A number of systems to support the construction of such mappings have recently been developed. A generic and effective benchmark for reliable and comparable evaluation of the practical utility of such systems would make an important contribution to the development of ontology-based data integration systems and their application in practice. We have proposed such a benchmark, called RODI. In this paper, we present a new version of RODI, which significantly extends our previous benchmark, and we evaluate various systems with it. RODI includes test scenarios from the domains of scientific conferences, geographical data, and oil and gas exploration. Scenarios are constituted of databases, ontologies, and queries to test expected results. Systems that compute relational-to-ontology mappings can be evaluated using RODI by checking how well they can handle various features of relational schemas and ontologies, and how well the computed mappings work for query answering. Using RODI, we conducted a comprehensive evaluation of seven systems.

Semantic access to streaming and static data at Siemens

“…The Optique platform allows to: create and edit mappings; create, edit, and import ontologies [54][55][56][57][58][59]; integrate several relational databases and data streams; formulate and visualise one time and continuous queries; efficiently process both static and streaming information; and browse query results. Thus, the Optique platform satisfies all the Siemens system Requirements R1-R5.…”

Ontology-Based Data Access Mapping Generation Using Data, Schema, Query, and Mapping Knowledge

“…Nevertheless, these RDB tools still fall short when tackling more complex challenges, e.g., when subclasses are grouped in a single table and need to be separated, and in real-life scenarios with more complex queries [2,17].…”

“…With existing knowledge, we refer to knowledge that is available before the mapping process. Although these tools are able to generate a promising mapping for simple scenarios, they fail on scenarios with more complex data sources involving non-trivial data schemas (hereinafter referred to as complex challenges) [2].…”

“…However, Linked Data applications have the need to access data that is available in non-rdf formats [2]. Ontology-Based Data Access (OBDA) systems provides such access where an ontology mediates between the raw data and its consumers [6].…”

“…Subsequently, the mapping process requires a low user effort. In most cases, these tools only use the original data schema and ignore other knowledge available in existing (Linked) Data [7,8], Schemas (data schemas, ontologies and vocabularies), the Queries that will be executed on the new rdf data (query workload) [2], and Mappings [9] (DSQM). With existing knowledge, we refer to knowledge that is available before the mapping process.…”