Semantic Annotation of Web of Things Using Entity Linking

Abstract

The web of things (WoT) improves syntactic interoperability between internet of things (IoT) devices by leveraging web standards. However, the lack of a unified WoT data model remains a challenge for the semantic interoperability. Fortunately, semantic web technologies are taking this challenge over by offering numerous semantic vocabularies like the semantic sensor networks (SSN) ontology. Although it enables the semantic interoperability between heterogeneous devices, the manual annotation hinders the scalability of the WoT. As a result, the automation of the semantic annotation of WoT devices becomes a prior issue for researchers. This paper proposes a method to improve the semi-automatic semantic annotation of web of things (WoT) using the entity linking task and the well-known ontologies, mainly the SSN.

SiSEG-Auto Semantic Annotation Service to Integrate Smart Energy Data

“…For this purpose, one of the techniques that can be employed is Word2vec. Word2vec provides a numerical representation of each distinct word in the form of a vector [40]. By comparing Word2vec vectors of keywords in the input with the labels of URIs available in knowledge bases [41], a single float number is obtained which can be appended into the already defined Feature Vector.…”