Research on the tag aided IoT service clustering

IoT Service clustering is the basis for service discovery, service composition and service recommendation in Internet of Things. However, most existing IoT service clustering approaches solely focus on utilizing service description text. This paper proposes a Tag Aided Service Clustering approach (TASC) which integrates both service description text and user-contributed tag information through Tag Aided Latent Dirichlet Allocation for IoT. Meanwhile, in order to improve the accuracy of latent topic extracted by topic model, we convert short text theme modeling to long text theme modeling by enriching the service description text semantically. The experimental results in which precision and recall were used as the evaluation metric demonstrate that our method is better than the existing methods.