Towards Service Ontology for Web Services Storage and Discovery

Q-KGSWS: Querying the Hybrid Framework of Knowledge Graph and Semantic Web Services for Service Discovery

“…Furthermore, unlike existing methods [20] that create matchmaking ontologies for discovery, our approach does not require such intermediary steps. For validation purposes, our work utilized the relevance file provided for the OWLS dataset in Phase 3 instead of employing alternative methods [29][30][31][32][33] to find relevant services, thereby enhancing the reliability of the framework. Additionally, we considered the complete OWLS-TC dataset rather than using its subset to generate KGSWS and further for service discovery.…”

“…In contrast, our approach, which seamlessly integrates the advantages of in depth querying over Knowledge Graph, overcame this bottleneck, leading to a more streamlined and efficient service discovery process. An important observation in our results was that while some subsequent studies [29][30][31][32] explored machine learning-based classification techniques, none had harnessed the potential of the Knowledge Graph within this domain.…”

“…However, none of these works implemented the concept of KGs in the domain of SWS. The work cited in study [31] [32] employed machine learning methodologies; specifically K-Means clustering was used in [31] to generate a Service Ontology from SWS corpus. Results demonstrated that the generated ontology can be used for the discovery mechanism.…”

“…However, with advanced machine learning methods, the overall accuracy of SWS discovery has improved. However, most of these works [29][30][31][32] focused on a subset of domains within the OWLS collection, and none of the works have implemented an automatic querying method for user queries. Instead of using the relevance file provided for OWLS-TC as per user queries, different methods have been employed to compute evaluation metrics.…”