1 January 2021 in IEEE Internet of Things Journal
The next-generation network (6G) has more strict requirements for the online learning ability and high interpretability of the learned systems. Machine learning is expected to be essential to assist in making the networks efficient and adaptable, but most promising methods often are treated as “black boxes” due to the deep structures and high nonlinearity. Therefore, this article attempts to study the interpretations of machine learning algorithms to make them more applicable to the 6G-enabled Internet of things (IoT) networks. Typically, this article focuses on the new model GBDT2NN, which distills the knowledge learned by gradient boosting decision tree (GBDT) into neural network (NN) models to retain the learning ability of numerical data and rise the ability of online learning at the same time, but it loses the interpretability. This article conducts an empirical study on explaining individual prediction of GBDT2NN by taking use of the feature importance learned from GBDT, and then further explores whether the explanation can improve the approximation process. In addition, this article proposes two methods to obtain the interpretations: 1) the independent method and 2) the joint method. The experiments on several data sets of IoT networks show that the proposed methods can achieve better performance on both explanations and predictions.
16 February 2018 in Journal of Ambient Intelligence and Humanized Computing
23 March 2019 in Multimedia Tools and Applications
1 January 2019 in International Journal of Embedded Systems
1 January 2017 in International Journal of High Performance Systems Architecture
Forests are precious and indispensable natural resource. Using wireless sensor networks (WSN) to establish forest fire prevention system can effectively reduce loss of human and economy. However, monitoring a very large and complex area such as the entire forest by WSN would entail a large number of sensor nodes, which are hard to be replaced or recharged. Such working conditions and constrains give rise to the necessity to make careful design on network protocol for the WSN. In this paper, we compare and analyse several key parameters for the location-based routing protocols and hierarchical routing protocols. This research will assist the users to design or select the best routing protocols for their applications.