Ontology Learning for Cost-Effective Large-Scale Semantic Annotation of Web Service Interfaces

Evaluation of a Semi-automated Semantic Annotation Approach for Bootstrapping the Analysis of Large-Scale Web Service Networks

“…In other words, when a given part/element name is assigned to an imprecise concept, then designated matching rules can not compensate this deficiency. In our work, accuracy of matching is mostly affected by performance of our ontology learning steps [6] which in turn, partially depends on the syntactic (e.g. spelling,…) and semantic (e.g.…”

“…For annotation of services in both datasets, we create two independent reference domain ontologies. While the first domain ontology is constructed manually by an ontology engineer, the second one is constructed automatically using our ontology learning mechanism [6]. We refer to the former and the latter respectively as the golden and the generated ontology.…”

“…AUTOMATED WEB SERVICE ANNOTATION Our automated annotation method combines a previously proposed ontology learning method [6] and a semi-automatic semantic annotation method [1] whereas the former provides input to the latter in form of annotation heuristics. The learning method is used to generate reference ontology from a corpus of Web service descriptions and then utilize the generated reference ontology to generate annotation heuristics.…”

“…The framework is applied to a novel semi-automatic cost-effective semantic annotation method, which is introduced as well. The latter combines a previously proposed ontology learning method [6] and a cost-effective semiautomatic semantic annotation method [1], for bootstrapping analysis of large repositories of Web services currently without semantic annotations. The annotation method starts with semi-automatically generating reference ontology.…”

Trust and Privacy Enabled Service Composition Using Social Experience

“…For newly introduced layers, we point out relevant sections in this paper while for some other layers references to our previous works are provided: A) Requirement Expansion Layer: It expands user requirement statement, specified in terms of available input and expected output parameters of services, with relevant concepts in order to increase service discovery efficiency. We obtain these terms and concepts from our pre-populated knowledge base which is built based on our ontology learning methodology [22]. The requirement expansion is performed according to the method proposed by Kungas and Dumas [11].…”