Discovery of IoT service based on probabilistic theme model

Discovery of IoT service based on probabilistic theme model

Weijiang Jinzhi Xu Yu

For the mass of the number of services in the Internet of things, according to the heterogeneity of service description and the resource height limitation and mobility of the equipment service, this paper presents a method of discovering the Internet of Things based on probabilistic theme model. The main features of this method are: 1 The use of English Wikipedia to build a high-quality theme model, and the semantic extension of the service text description of similar short texts, the theme model can be used to estimate the implied theme of service text description more effectively; 2 The paper proposes to study the implicit theme of service text using nonparametric subject model, and reduce the model training time; 3 The service is automatically classified and text similarity calculated by using the implied theme of the service, Quickly reduce the number of service matching, speed up service text similarity calculation; 4 to support signature of wsdl-based and restful two kinds of IoT services Matching algorithm. The experimental results show that the accuracy of this method (precision) and the normalized impairment cumulative gain (NDCG) are greatly improved compared with the existing method of IoT service discovery.

Discovery of IoT service based on probabilistic theme model