No abstract

The ocean engineering equipment industry is the foundation for the implementation of maritime strategy. China’s national departments at all levels have developed relevant ocean engineering equipment industry policies to promote the rapid development of the industry. By using 56 industrial policies issued between 2010 and 2020 as the research sample, we conducted an in-depth assessment of the external structural characteristics and structure of the main cooperation network for such policies using descriptive statistics and social network analysis. Based on a symmetric analysis method, the two-dimensional matrix of cooperation breadth and cooperation depth, together with the measurement of the issuing subject’s centrality, was used to analyze the evolution of the subject’s role in the network. The research shows that the development of China’s ocean engineering equipment industry policies can be divided into three stages, and there are the following problems during the development of policies: (1) some policies and regulations are imperfect; (2) the network of cooperation among joint issuers is limited; and (3) some policies are issued by multiple government departments, but there is a lack of specialized and unified management from an absolute core department. Based on the above problems, we present some suggestions for policy optimization at the end of this paper.

Homogeneous cross-project defect prediction (HCPDP) aims to apply a binary classification model built on source projects to a target project with the same metrics. However, there is still room for improvement in the performance of the existing HCPDP models. This study has proposed a novel approach, including one-to-one and many-to-one predictions. First, we apply the Jensen-Shannon divergence to select the most similar source project automatically. Second, relative density estimation is introduced to choose the suitable instance of the selected source project. Third, one-to-one and many-to-one prediction models are trained by the selected instances. Finally, two benchmark datasets are used to evaluate the proposed approach. Compared to the state-of-the-art methods, the experimental results demonstrated that the proposed approach could improve the prediction performance in the F1-score, AUC, and G-mean metrics and exhibit strong adaptability to the traditional classifiers.