Yong Shi

With the explosion of social media networks, many modern applications are concerning about people's connections, which leads to the so-called social computing. An elusive question is to study how opinion communities form and evolve in real-world networks with great individual diversity and complex human connections. In this paper, we attempt to model a realistic social media network as a discrete-time dynamical system, where the opinion matrix and the community structure could mutually affect each other. The community detection in social media networks is naturally formulated as a multi-objective optimization problem, i.e., finding a set of densely connected components with similar opinion vectors. We propose a novel and powerful graph K-means framework, which is composed of three coupled phases in each discrete-time period. Specifically, the first phase uses a fast heuristic approach to identify those opinion leaders who have relatively high local reputation; the second phase adopts a novel dynamic game model to find the locally Pareto-optimal community structure; the final phase employs a robust opinion dynamics model to simulate the evolution of the opinion matrix. We conduct a series of comprehensive experiments on real-world benchmark networks to validate the performance of GK-means through comparisons with the state-of-the-art graph clustering technologies.

In classification based on multiple-criteria linear programming (MCLP), we need to find the optimal solution of the MCLP problem as a classifier. According to dual theory, multiple criteria can be switched to multiple constraint levels, and vice versa. A MCLP problem can be logically extended into a multiple-criteria and multiple-constraint levels linear programming (MC2LP) problem. In many applications, such as credit card account classification, how to handle two types of error is a key issue. The errors can be caused by a fixed cutoff between a "Good" group and a "Bad" group. Two types of error can be systematically corrected by using the structure of MC2LP, which allows two alterable cutoffs. In order to do so, a penalty (or cost) is imposed to find the potential solution for all possible trade-offs in solving MC2LP problem. Some correction strategies can be investigated by the solution procedure. Furthermore, a framework of decision supporting system can be illustrated for various real-life applications of the proposed method.

Classification problem has attracted an increasing amount of interest. Various classifiers have been proposed in the last decade, such as ANNs, LDA, and SVM. Regular Multiple Criteria Linear Programming (RMCLP) is an effective classification method, which was proposed by Shi and his colleagues and have been applied to handle different real-life data mining problems. In this paper, inspired by the application potential of RMCLP, we propose a novel Structural RMCLP (called SRMCLP) method for classification problem. Unlike RMCLP, SRMCLP is sensitive to the structure of the data distribution and can construct more reasonable classifiers by exploiting these prior data distribution information within classes. The corresponding optimization problem of SRMCLP can be solved by a standard quadratic programming. The effectiveness of the proposed method is demonstrated via experiments on synthetic and available benchmark datasets.

Timely response is extremely important in emergency management. However, cardinal inconsistent data may exist in a judgment matrix because of the limited expertise, preference conflict as well as the complexity nature of the decision problems. The existing inconsistent data processing models for positive reciprocal matrix either are complicated or dependent on the priority weights, which will delay the decision making process in emergency. In this paper, a geometric mean induced bias matrix (GMIBM), which is only based on the original matrix A, is proposed to quickly identify the most inconsistent data in the judgment matrix. The correctness and effectiveness of the proposed model are proved mathematically and illustrated by two numerical examples. The results show that the proposed model not only preserves most of the original information in matrix A, but also is faster than existing methods.

This paper proposes two-stage hybrid feature selection algorithms to build the stable and efficient diagnostic models where a new accuracy measure is intro