Civil Aviation University of China

About: Civil Aviation University of China is a education organization based out in Tianjin, China. It is known for research contribution in the topics: Air traffic control & Civil aviation. The organization has 5647 authors who have published 4559 publications receiving 29825 citations. The organization is also known as: Zhōngguó Mínháng Dàxué.

Carrot/stick mechanisms for collection responsibility sharing in multi-tier closed-loop supply chain management

Abstract: This paper studies the government’s carrot/stick (reward-penalty) mechanisms for the core collection responsibility in a multi-tier closed-loop supply chain (CLSC). Within a Stackelberg game with the manufacturer as the leader, our study identifies the optimal reward-penalty mechanisms (hereafter RPMs) in three sharing modes of responsibility for the used product collection: (i) a CLSC without the RPM, (ii) a CLSC with the RPM imposed on the manufacturer only, and (iii) a CLSC with the RPM enforced on both the manufacturer and the retailer. From a collection rate and social welfare perspective, we find that it is optimal to put the collection responsibility on the manufacturer only, and the government should force a relatively high reward-penalty intensity on the manufacturer. Although the mode with the RPM imposed only on the manufacturer results in the highest buy-back price and the lowest wholesale price, this is the social responsibility the manufacturer must bear to benefit the environment. Finally, the numerical results show that the collection rate rises as the reward-penalty intensity and the sharing ratio of the manufacturer. Besides, high reward-penalty intensity and high sharing ratio are conducive to improving social welfare.

MIKE: Keyphrase Extraction by Integrating Multidimensional Information

Abstract: Traditional supervised keyphrase extraction models depend on the features of labelled keyphrases while prevailing unsupervised models mainly rely on structure of the word graph, with candidate words as nodes and edges capturing the co-occurrence information between words. However, systematically integrating all these multidimensional heterogeneous information into a unified model is relatively unexplored. In this paper, we focus on how to effectively exploit multidimensional information to improve the keyphrase extraction performance (MIKE). Specifically, we propose a random-walk parametric model, MIKE, that learns the latent representation for a candidate keyphrase that captures the mutual influences among all information, and simultaneously optimizes the parameters and ranking scores of candidates in the word graph. We use the gradient-descent algorithm to optimize our model and show the comprehensive experiments with two publicly-available WWW and KDD datasets in Computer Science. Experimental results demonstrate that our approach significantly outperforms the state-of-the-art graph-based keyphrase extraction approaches.