A web service selection model based on DeepFM and hash space segment matching

QoS is an important indicator to measure the quality of web services. How to choose the most suitable web service is an urgent problem to solve. Factors such as independent QoS attributes and data sparseness may cause poor service selection accuracy, making it impossible to select matching web services for users. Therefore, this paper proposes an algorithm for service selection based on DeepFM model and hash list. By learning the interaction relationship between low-order features and high-order features of QoS attributes based on the DeepFM model and DeepFM model can effectively solve the cold start problem caused by data sparseness. To improve the accuracy of QoS prediction, the hash-based service matching algorithm is used to quickly select service resources. Experiments show that the proposed model is helpful to improve the effect of service selection.