Application of the Swarm Intelligence for the Semantic Annotation of the Web of Things
International Journal of Advanced Trends in Computer Science and Engineering · 2020
Entity Linking (EL) is an important tool for many researches including the Internet of Things (IoT).Particularly, it helps to automate the annotation of the IoT devices and services with semantic vocabulary.This semantic annotation improves the interoperability between IoT devices during services discovery and composition.This paper describes an ELbased semantic annotation framework.The latter processes a web table to extract the mentions, generates candidate entities from the Knowledge base (KB) for each mention and then disambiguates the mentions to the correct entry in the KB.The main contribution of this paper is the use of both genetic algorithm (GA) and Artificial Bee Colony (ABC) to handle the combinatorial problem of the collective disambiguation.The proposed framework implements, in addition, some text preprocessing techniques such as the bag of words (BoW) model and provides a tool for IoT data collection.The framework is a first step to demonstrate how swarm intelligence (SI) can exceed the numerical approaches and even some meta-heuristic techniques like the probabilistic based ones.