Separating Latent Classes by Information Criteria

This study evaluates performance of information criteria used to separate latent classes. In the evaluations, various numbers of latent classes, sample sizes, parameter structures and latent-class complexities were designed to simulate datasets. The average accuracy rates of information criteria in selecting the designed numbers of latent classes were the core results in this experiment. The study revealed that widely used information criteria, e.g., AIC, BIC, CAIC, could perform poorly under some circumstances. By including a sample size adjustment (Rissanen, 1978), the unsatis-factory performances could be improved considerably. The sample size adjustment provides a plausible solution for separating latent classes. Guidelines are provided to help achieve optimum use of the model fit indices.     

扫描检测平台效用评估

利用网络中未使用IP地址构建扫描检测平台能够有效提高检测准确率,降低虚警率.在扫描检测平台实际构建之前,需要对可控部署地址资源的预计检测效用进行评估,以确定地址资源可达的检测目标以及明确平台构建的必要性.为此,提出一种基于网络分类的扫描检测模型,依据此模型可对扫描检测平台对于随机扫描源和本地优先扫描源的检测效用进行评估,为扫描检测平台部署构建提供理论指导.以法国电信Leurre’com Honeynet Project实际分布式检测平台作为效用评估样例.评估结果表明该检测平台能够有效检测类似Slammer蠕虫的高速随机扫描源和每秒至少发出2个扫描连接的本地优先扫描源,对低速扫描源检测效用低下.实际数据统计结果与仿真实验验证了评估结果的准确性.     

一种新的基于公司价值的可转债定价方法

可转债近年来逐渐受到市场的关注,其定价方法也成为理论界研究的重点.传统的可转债定价方法的基本思路是通过建立可转债价值的模型来直接求解可转债理论价格.试图建立一种新的基于公司价值的可转债定价模型,希望通过分析可转债、股票和公司价值三者之间的关系,建立一个关于公司价值的方程,然后通过求解公司价值间接地得到可转债价值的方法,并分析了该方法的理论缺陷.最后,对招商转债进行了实证研究,并将其与市场价格进行比较,指出了理论价格被高估的原因.     

A situation calculus based approach to dynamic management of e-business services

In the era of e-businesses, the traditional business services are greatly challenged by the ever-increasing demands from customers with various backgrounds and personalities. Large numbers of new e-businesses are driven by the needs of customers. The existing dynamics inherent in the customer needs require the corresponding dynamic management of services. Attempting to respond to customers in a rapid and intelligent way, this paper proposes a situation calculus based approach for dynamically managing e-Business services in the ubiquitous environment. By employing the formalism of the situation calculus to enable intelligence and automation, the approach can implement. the functions of service automatic composition and model verification. These functions will improve the degree of customer-orientation and enable fast responsiveness in the emerging e-service systems.     

Conflict analysis approaches for investigating attitudes and misperceptions in the War of 1812

Formal systems engineering approaches to modeling misperceptions and attitudes are employed within the framework of the graph model for conflict resolution to systematically study the War of 1812 between the United States of America and Great Britain in order to provide enhanced insights into the causes of the war. More specifically, relational definitions for preferences, movements and stability concepts are defined for describing the attitudes and associated behavior of decision makers involved in a conflict. To capture misperceptions of decision makers in the War of 1812, attitudes are studied within the structure of a hypergame. Combining attitudes and misperceptions within the paradigm of the graph model furnishes the flexible analytical tool which demonstrates that misunderstanding of attitudes by Great Britain and the United States may have contributed to the outbreak of this nasty war.     

THE λ-GROBNER BASES UNDER POLYNOMIAL COMPOSITION

Polynomial composition is the operation of replacing variables in a polynomial with other polynomials. λ-Grgbner basis is an especial Grobner basis. The main problem in the paper is： when does composition commute with λ-Grobner basis computation？ We shall answer better the above question. This has a natural application in the computation of λ-Grobner bases.     

A multiple intelligent agent system for credit risk prediction via an optimization of localized generalization error with diversity

Company bankruptcies cost billions of dollars in losses to banks each year. Thus credit risk prediction is a critical part of a bank＇s loan approval decision process. Traditional financial models for credit risk prediction are no longer adequate for describing today＇s complex relationship between the financial health and potential bankruptcy of a company. In this work, a multiple classifier system （embedded in a multiple intelligent agent system） is proposed to predict the financial health of a company. In our model, each individual agent （classifier） makes a prediction on the likelihood of credit risk based on only partial information of the company. Each of the agents is an expert, but has limited knowledge （represented by features） about the company. The decisions of all agents are combined together to form a final credit risk prediction. Experiments show that our model out-performs other existing methods using the benchmarking Compustat American Corporations dataset.     

Using Action Research to Improve Communication and Facilitate Group Work in a Multicultural Classroom: A South African Case Study

This study started with problems that my colleagues (and I) were experiencing with interaction and intercultural communication among students in our diverse classrooms. Educators were also experiencing difficulties in motivating students to work effectively in groups. The purpose of this paper was to seek solutions to these problems by exploring variations of the group oral assessment structure in a multilingual and multicultural context. Four phases of assessments were conducted at two tertiary institutions in South Africa using a combination of action research and a participatory approach. Not only did this study succeed in addressing the problems but it also revealed the added benefits of using an action research methodology.     

EXPLORATION/EXPLOITATION TRADEOFF WITH CELL-SHIFT AND HEURISTIC CROSSOVER FOR EVOLUTIONARY ALGORITHMS

In order to tradeoff exploration/exploitation and inspired by cell genetic algorithm a cellshift crossover operator for evolutionary algorithm （EA） is proposed in this paper. The definition domain is divided into n-dimension cubic sub-domains （cell） and each individual locates at an ndimensional cube. Cell-shift crossover first exchanges the cell numbers of the crossover pair if they are in the different cells （exploration） and subsequently shift the first individual from its initial place to the other individual＇s cell place. If they are already in the same cell heuristic crossover （exploitation） is used. Cell-shift/heuristic crossover adaptively executes exploration/exploitation search with the vary of genetic diversity. The cell-shift EA has excellent performance in terms of efficiency and efficacy on ten usually used optimization benchmarks when comparing with the recent well-known FEP evolutionary algorithm.     

INFORMATION CHARACTERIZATION OF COMMUNICATION CHANNELS FOR SYSTEM IDENTIFICATION

This paper studies identification of systems in which the system output is quantized, transmitted through a digital communication channel, and observed afterwards. The concept of the CR Ratio is introduced to characterize impact of communication channels on identification. The relationship between the CR Ratio and Shannon channel capacity is discussed. Identification algorithms are further developed when the channel error probability is unknown.