No abstract

No abstract

Abstract-In Dempster-Shafer evidence theory (DST) based classifier design, Dempster's combination (DC) rule is commonly used as a multi-attribute classifier to combine evidence collected from different attributes. The main aim of this paper is to present a classification method using a novel combination rule i.e., the evidence reasoning (ER) rule. As an improvement of the DC rule, the newly proposed ER rule defines the reliability and weight of evidence. The former indicates the ability of attribute or its evidence to provide correct assessment for classification problem, and the latter reflects the relative important of evidence in comparison with other evidence when they need to be combined. The ER rule-based classification procedure is expatiated from evidence acquisition and estimation of evidence reliability and weight to combination of evidence. It is a purely data-driven approach without making any assumptions about the relationships between attributes and class memberships, and the specific statistic distributions of attribute data. Experiential results on five popular benchmark databases taken from University of California Irvine (UCI) machine learning database show high classification accuracy that is competitive with other classical and mainstream classifiers.

No abstract

Research project evaluation and selection is mainly concerned with evaluating a number of research projects and then choosing some of them for implementation. It involves a complex multiple-experts multiple-criteria decision making process. Thus this paper presents an effective method for evaluating and selecting research projects by using the recently-developed evidential reasoning (ER) rule. The proposed ER rule based evaluation and selection method mainly includes (1) using belief structures to represent peer review information provided by multiple experts, (2) employing a confusion matrix for generating experts' reliabilities, (3) implementing utility based information transformation to handle qualitative evaluation criteria with different evaluation grades, and (4) aggregating multiple experts' evaluation information on multiple criteria using the ER rule. An experimental study on the evaluation and selection of research proposals submitted to the National Science Foundation of China demonstrates the applicability and effectiveness of the proposed method. The results show that (1) the ER rule based method can provide consistent and informative support to make informed decisions, and (2) the reliabilities of the review information provided by different experts should be taken into account in a rational research project evaluation and selection process, as they have a significant influence to the selection of eligible projects for panel review.

Please cite this article in press as: Y.-W. Chen et al., On the inference and approximation properties of belief rule based systems, Inform. Sci.

No abstract

No abstract