Semantic Rules for Service Discovery in Social Internet of Things

World is moving towards preceding generation Alpha that believes in socializing the objects or devices within the network known as Social Internet of Things(SIoT). In SIoT the services are discovered based on the user preferences by establishing relationship between the objects and user. Thus object intelligent model is necessary to efficiently discover the services based on the context and semantic relationships between the users and objects. SIoT posses certain challenges like service discovery, compositions, etc. To address these challenges, in this work a semantic rules are proposed to discover services and it is evaluated using various machine learning algorithms on SIoT data. Hence, this work focusses on implementing and evaluating the various machine learning algorithms on SIoT device data considering health as an application. The performance is evaluated with dataset ratio of 60:40, 70:30, 80:20 and 90:10. The machine learning classifier techniques like Decision Tree, Naive Bayes, K-nearest neighbour and Artificial Neural Network classifiers gives the average accuracy 100%, 99%, 100%and 100% respectively. The results show that ANN gives best results compared to other algorithms.