Diversity measures for multiple classifier system analysis and design

In the context of Multiple Classifier Systems, diversity among base classifiers is known to be a necessary condition for improvement in ensemble performance. In this paper the ability of several pair-wise diversity measures to predict generalisation error is compared. A new pair-wise measure, which is computed between pairs of patterns rather than pairs of classifiers, is also proposed for two-class problems. It is shown experimentally that the proposed measure is well correlated with base classifier test error as base classifier complexity is systematically varied. However, correlation with unity-weighted sum and vote is shown to be weaker, demonstrating the difficulty in choosing base classifier complexity for optimal fusion. An alternative strategy based on weighted combination is also investigated and shown to be less sensitive to number of training epochs.     

An analysis of diversity measures

Diversity among the base classifiers is deemed to be important when constructing a classifier ensemble. Numerous algorithms have been proposed to construct a good classifier ensemble by seeking both the accuracy of the base classifiers and the diversity among them. However, there is no generally accepted definition of diversity, and measuring the diversity explicitly is very difficult. Although researchers have designed several experimental studies to compare different diversity measures, usually confusing results were observed. In this paper, we present a theoretical analysis on six existing diversity measures (namely disagreement measure, double fault measure, KW variance, inter-rater agreement, generalized diversity and measure of difficulty), show underlying relationships between them, and relate them to the concept of margin, which is more explicitly related to the success of ensemble learning algorithms. We illustrate why confusing experimental results were observed and show that the discussed diversity measures are naturally ineffective. Our analysis provides a deeper understanding of the concept of diversity, and hence can help design better ensemble learning algorithms. Editor: Tom Fawcett     

模糊粗糙集测度的分析分层建模及应急演练控制应用

针对应急救援演练控制的复杂性和难以量化问题,为实现多人参演系统的有效控制,基于分析分层过程法(analytic hierarchy process,AHP),建立一种模糊粗糙集知识测度的综合建模方法.首先,分析模糊粗糙集各类知识测度相关概念、相互联系和各自特点,通过AHP方法,建立模糊规则的分层度量模型并给出了对比矩阵的构造示例,对模糊规则进行更加精细的度量.其次,在分析应急演练知识构成的基础上,提出预案知识提取和模糊关系粗糙集的构建方法;设计了演练过程控制流程和基于本文知识综合测度方法形成的核心控制流程;通过对规则重要性排序,提高规则判别精度,提供规则选择的手段和一种规则冲突消解方法.最终,通过一个简单案例,验证了本文所提的研究工作的可行性.     

Mapping vegetation and land cover in a large urban area using a multiple classifier system

Vegetation and land-cover information is critical for sustainable environmental management in urban areas. Remote sensing has increasingly been used to derive such information, yet it has been challenged by the spectral and spatial complexity in the urban environment. In this study, we developed a multiple classifier system (MCS) to help improve remote-sensing-based vegetation and land-cover mapping in a large metropolitan area. MCSs, although considered as an emerging hot topic and a promising trend in pattern recognition, have not received the attention it deserves in the remote-sensing community. Our work consisted of several components. First, we identified a group of commonly used pattern recognizers from different families of statistical learning algorithms as base classifiers. Then, we implemented them to derive land-cover information from a satellite image covering the study site. Last, we adopted a weighting and combination method to generate the final map. Results indicate that there is statistically significant difference in the classification accuracy between the MCS developed and each base classifier considered. Comparing with the base classifiers, the MCS produced not only about 5–8% higher overall classification accuracy but also the most stable categorical accuracies. Moreover, the MCS generated a larger accuracy improvement for spectrally complex classes than for relatively homogenous ones, suggesting its comparative advantage in reducing classification errors caused by class ambiguity. The novelties of our work are with the demonstration of how MCSs can be operationally used to improve image classification from large remote sensor data sets with complex patterns and with the insight into the behaviour of MCSs in relation to the complexity of individual classes. These findings can help promote the use of MCSs as an emerging premier approach for image classification by the remote-sensing community.     

Performance evaluation of rule‐based expert systems: An example from medical billing domain

Most of the research in the area of performance evaluation of rule‐based expert systems (RBESs) is focused on verification and validation issues. Many researchers discuss usability, usefulness, portability, and response time for the evolution of RBES. The final goal of all such studies is to construct a system with optimal, accurate knowledge base. Arguably, a system with best knowledge base is actually worthless if it is never utilized in the real world. We have proposed “benefit” as a measure of evaluation and suggested some guidelines for performance evaluation of RBESs. The proposed measure has been demonstrated for performance evaluation of an RBES applied in the medical billing domain. Results showed that the system has saved hundreds of working hours during the evaluation period of 3 months. Moreover, other associated measures have also been considered. Associated measures in the medical billing domain are “claim rejection rate”—reduced by 54%—and “claim aging,” which has decreased from 34 to 28 days due to the RBES. Guidelines proposed by this research can be applied for the evaluation of expert systems implemented in other application domains, including in the first place business decision support systems.     

On the measure of conflicts: Shapley Inconsistency Values

There are relatively few proposals for inconsistency measures for propositional belief bases. However inconsistency measures are potentially as important as information measures for artificial intelligence, and more generally for computer science. In particular, they can be useful to define various operators for belief revision, belief merging, and negotiation. The measures that have been proposed so far can be split into two classes. The first class of measures takes into account the number of formulae required to produce an inconsistency: the more formulae required to produce an inconsistency, the less inconsistent the base. The second class takes into account the proportion of the language that is affected by the inconsistency: the more propositional variables affected, the more inconsistent the base. Both approaches are sensible, but there is no proposal for combining them. We address this need in this paper: our proposal takes into account both the number of variables affected by the inconsistency and the distribution of the inconsistency among the formulae of the base. Our idea is to use existing inconsistency measures in order to define a game in coalitional form, and then to use the Shapley value to obtain an inconsistency measure that indicates the responsibility/contribution of each formula to the overall inconsistency in the base. This allows us to provide a more reliable image of the belief base and of the inconsistency in it.     

New similarity measures of intuitionistic fuzzy sets based on the Jaccard index with its application to clustering

A similarity measure is a useful tool for determining the similarity between two objects. Although there are many different similarity measures among the intuitionistic fuzzy sets (IFSs) proposed in the literature, the Jaccard index has yet to be considered as way to define them. The Jaccard index is a statistic used for comparing the similarity and diversity of sample sets. In this study, we propose a new similarity measure for IFSs induced by the Jaccard index. According to our results, proposed similarity measures between IFSs based on the Jaccard index present better properties. Several examples are used to compare the proposed approach with several existing methods. Numerical results show that the proposed measures are more reasonable than these existing measures. On the other hand, measuring the similarity between IFSs is also important in clustering. Thus, we also propose a clustering procedure by combining the proposed similarity measure with a robust clustering method for analyzing IFS data sets. We also compare the proposed clustering procedure with two clustering methods for IFS data sets.     

方向测度及其在纹理识别中的应用

在一均匀的纹理图象中沿任何一个方向象素的灰度变化都服从一定的规律.依据这个特 点,本文提出一种新的纹理统计特征--方向测度.方向测度能够提取纹理的二阶及高阶统 计特征.高阶方向测度对自然纹理具有很高的识别率.另外方向测度的计算量小,运算结构简 单,利于多机并行处理.     

Fingerprint Retrieval for Identification

This paper presents a front-end filtering algorithm for fingerprint identification, which uses orientation field and dominant ridge distance as retrieval features. We propose a new distance measure that better quantifies the similarity evaluation between two orientation fields than the conventional Euclidean and Manhattan distance measures. Furthermore, fingerprints in the data base are clustered to facilitate a fast retrieval process that avoids exhaustive comparisons of an input fingerprint with all fingerprints in the data base. This makes the proposed approach applicable to large databases. Experimental results on the National Institute of Standards and Technology data base-4 show consistent better retrieval performance of the proposed approach compared to other continuous and exclusive fingerprint classification methods as well as minutia-based indexing schemes     

A complexity measure for data flow models

For common data flow schemes, the number of copies of tokens made during a computation is shown to be a Blum complexity measure.⁽¹⁾ Results from abstract complexity theory (see Ref. 2) then hold for the copy measure, indicating, for example, that any implementation of a data flow processor will be constrained by its ability to copy tokens. The copy measure is a natural measure of complexity for data flow computations, and is distinct from the usual time or space measures. The result is generalized to a wider class of data flow schemas, including those with an apply operator. An example is also given of a data flow scheme which makes no copies.Supported in part by NSF Grant MCS 83-01536 and NSA OCREAE Grant MDA904-85-H-0002.     

集成学习中的多样性度量

在集成学习中, 基分类器之间的多样性对于解释多分类器系统的工作机理和构造有效的集成系统具有重要的作用, 但至今仍没有统一的度量多样性的方法. 首先总结介绍常用的多样性度量方法, 阐述每种方法评估多样性的角度和方式; 然后从对多样性新的解释和度量、多样性度量在选择性集成中的应用、多样性度量和集成学习精度的关系3 个方面探讨多样性度量的研究进展; 最后给出关于多样性度量进一步的研究方向.

     

色噪声激励的时滞非线性系统瞬态响应研究

建立了色噪声与时滞联合作用的非线性系统模型,提出求解其瞬态概率密度的高效近似算法。利用等价变换将时滞系统简化为非时滞系统;通过线性化方法和随机平均原理得到原系统振幅过程的平均It?随机微分方程和相应的Fokker-Planck-Kolmogorov（FPK）方程。基于退化线性系统导出一组正交基,在该基空间内进行Galerkin变分得到近似瞬态概率密度。将该方法应用到受色噪声激励的双时滞Duffing-Van Der Pol振子得到理论解,采用蒙特卡罗模拟（MCS）验证理论解的正确性。分析了色噪声参数和时滞参数对瞬态响应的影响。研究结果表明：所提理论方法可有效求解受色噪声激励的时滞非线性系统的瞬态概率密度;算法求解效率高于MCS;色噪声和时滞均明显影响了系统瞬态响应。     

The effects of combining classifiers with the same training statistics using Bayesian decision rules

Multiple classifier systems (MCS) are attracting increasing interest in the field of pattern recognition and machine learning. Recently, MCS are also being introduced in the remote sensing field where the importance of classifier diversity for image classification problems has not been examined. In this article, Satellite Pour l'Observation de la Terre (SPOT) IV panchromatic and multispectral satellite images are classified into six land cover classes using five base classifiers: contextual classifier, k-nearest neighbour classifier, Mahalanobis classifier, maximum likelihood classifier and minimum distance classifier. The five base classifiers are trained with the same feature sets throughout the experiments and a posteriori probability, derived from the confusion matrix of these base classifiers, is applied to five Bayesian decision rules (product rule, sum rule, maximum rule, minimum rule and median rule) for constructing different combinations of classifier ensembles. The performance of these classifier ensembles is evaluated for overall accuracy and kappa statistics. Three statistical tests, the McNemar's test, the Cochran's Q test and the Looney's F-test, are used to examine the diversity of the classification results of the base classifiers compared to the results of the classifier ensembles. The experimental comparison reveals that (a) significant diversity amongst the base classifiers cannot enhance the performance of classifier ensembles; (b) accuracy improvement of classifier ensembles can only be found by using base classifiers with similar and low accuracy; (c) increasing the number of base classifiers cannot improve the overall accuracy of the MCS and (d) none of the Bayesian decision rules outperforms the others.     

模糊软集之间的相似性度量及应用

研究模糊软集的不确定度量问题,给出模糊软集的包含度、相似度公理化定义;基于模糊蕴含算子提出新的模糊软集包含度与相似度度量方法,该方法具有一定的普遍性,在某种程度上提供不同的模糊蕴含算子就可得到不同的包含度与相似度。基于新的相似度度量方法构造了一种决策方法并应用于金融企业流动性检测中。     

Vague值与Vague集的多测度相似度量

针对Vague值(集)相似度量问题,指出了采用单一测度构造Vague值(集)相似度量的不足,根据不同测度表现的不同相似性度量特点,提出了两种由距离测度、未知度测度和熵测度三种测度结合的Vague值(集)多测度相似度量,并给出了相应的定义和性质.多测度相似度量体现出了多特征度量相似性的特点,若距离与熵测度失效,未知度测度仍能发挥作用,而距离与熵测度度量结果的综合,则进一步提高了分辨力.实例验证了该多测度相似度量的有效性和优越性.     

多分类器融合系统设计与应用

设计高性能的分类系统是模式识别研究领域追求的目标,多分类器系统MCS是实现该目标的一个有效途径。在对比MCS与单分类器系统SCS的基础上,阐述了MCS的设计与优化,并对当前的优化技术进行了分类和比较,指出了存在的问题及未来的研究方向。给出了一个用MCS实现空战目标识别的应用实例,该实例以目标的战术性能参数为分类特征,通过和规则融合多个BP网络分类器求得系统决策。实验结果表明,MCS能显著提高系统的识别率和可信度。     

QoS Assessment of Mobile Crowdsensing Services

The wide spreading of smart devices drives to develop distributed applications of increasing complexity, attracting efforts from both research and business communities. Recently, a new volunteer contribution paradigm based on participatory and opportunistic sensing is affirming in the Internet of Things scenario: Mobile Crowdsensing (MCS). A typical MCS application considers smart devices as contributing sensors able to produce geolocalized data about the physical environment, then collected by a remote application server for processing. The growing interest on MCS allows to think about its possible exploitation in commercial context. This calls for adequate methods able to support MCS service providers in design choices, implementing mechanisms for the quality of service (QoS) assessment while dealing with complex time-dependent phenomena and churning issues due to contributors that unpredictably join and leave the MCS system. In this paper, we propose an analytical modeling framework based on stochastic Petri nets to evaluate QoS metrics of a class of MCS services. This method requires to extend the Petri net formalism by specifying a marking dependency semantics for non-exponentially distributed transitions. The approach is then applied to an MCS application example deriving some QoS measures that can drive quantitative evaluation and characterization of the “crowd” behavior.     

A general framework for measuring inconsistency through minimal inconsistent sets

Hunter and Konieczny explored the relationships between measures of inconsistency for a belief base and the minimal inconsistent subsets of that belief base in several of their papers. In particular, an inconsistency value termed MIV _C , defined from minimal inconsistent subsets, can be considered as a Shapley Inconsistency Value. Moreover, it can be axiomatized completely in terms of five simple axioms. MinInc, one of the five axioms, states that each minimal inconsistent set has the same amount of conflict. However, it conflicts with the intuition illustrated by the lottery paradox, which states that as the size of a minimal inconsistent belief base increases, the degree of inconsistency of that belief base becomes smaller. To address this, we present two kinds of revised inconsistency measures for a belief base from its minimal inconsistent subsets. Each of these measures considers the size of each minimal inconsistent subset as well as the number of minimal inconsistent subsets of a belief base. More specifically, we first present a vectorial measure to capture the inconsistency for a belief base, which is more discriminative than MIV _C . Then we present a family of weighted inconsistency measures based on the vectorial inconsistency measure, which allow us to capture the inconsistency for a belief base in terms of a single numerical value as usual. We also show that each of the two kinds of revised inconsistency measures can be considered as a particular Shapley Inconsistency Value, and can be axiomatically characterized by the corresponding revised axioms presented in this paper.     

一种新型多目标智能加权灰靶决策模型

提出一种新的多目标加权灰靶决策模型.针对具有满意域的效益型、成本型和适中型等不同性质的决策目标,构造出效益型目标效果测度、成本型目标效果测度、适中型目标下限效果测度、适中型目标上限效果测度等4种新型一致效果测度函数.据此可将具有不同意义、不同量纲、不同性质的决策目标转换为一致效果测度,从而能够方便地求出综合效果测度矩阵.将灰靶临界值设计为一致效果测度函数的正负分界点(即零点),充分考虑了目标效果值中靶和脱靶两种不同情形,物理涵义清晰,综合效果测度的分辨率也大大提高.