Heterogeneous Information Networks (HINs) contain multiple types of nodes and edges; therefore, they can preserve the semantic information and structure information. Cluster analysis using an HIN has obvious advantages over a transformation into a homogenous information network, which can promote the clustering results of different types of nodes. In our study, we applied a Nonnegative Matrix Tri-Factorization (NMTF) in a cluster analysis of multiple metapaths in HIN. Unlike the parameter estimation method of the probability distribution in previous studies, NMTF can obtain several dependent latent variables simultaneously, and each latent variable in NMTF is associated with the cluster of the corresponding node in the HIN. The method is suited to co-clustering leveraging multiple metapaths in HIN, because NMTF is employed for multiple nonnegative matrix factorizations simultaneously in our study. Experimental results on the real dataset show that the validity and correctness of our method, and the clustering result are better than that of the existing similar clustering algorithm.

The Internet of Things emphasizes the concept of objects connected with each other, which includes all kinds of wireless sensor networks. An important issue is to reduce the energy consumption in the sensor networks since sensor nodes always have energy constraints. Deployment of thousands of wireless sensors in an appropriate pattern will simultaneously satisfy the application requirements and reduce the sensor network energy consumption. This article deployed a number of sensor nodes to record temperature data. The data was then used to predict the temperatures of some of the sensor node using linear programming. The predictions were able to reduce the node sampling rate and to optimize the node deployment to reduce the sensor energy consumption. This method can compensate for the temporarily disabled nodes. The main objective is to design the objective function and determine the constraint condition for the linear programming. The result based on real experiments shows that this method successfully predicts the values of unknown sensor nodes and optimizes the node deployment. The sensor network energy consumption is also reduced by the optimized node deployment.