Chameleon based on clustering feature tree and its application in customer segmentation

Clustering analysis plays an important role in the filed of data mining. Nowadays, hierarchical clustering technique is becoming one of the most widely used clustering techniques. However, for most algorithms of hierarchical clustering technique, the requirements of high execution efficiency and high accuracy of clustering result cannot be met at the same time. After analyzing the advantages and disadvantages of the hierarchical algorithms, the paper puts forward a two-stage clustering algorithm, named Chameleon Based on Clustering Feature Tree (CBCFT), which hybridizes the Clustering Tree of algorithm BIRCH with algorithm CHAMELEON. By calculating the time complexity of CBCFT, the paper argues that the time complexity of CBCFT increases linearly with the number of data. By experimenting on sample data set, this paper demonstrates that CBCFT is able to identify clusters with large variance in size and shape and is robust to outliers. Moreover, the result of CBCFT is as similar as that of CHAMELEON, but CBCFT overcomes the shortcoming of the low execution efficiency of CHAMELEON. Although the execution time of CBCFT is longer than BIRCH, the clustering result of CBCFT is much satisfactory than that of BIRCH. Finally, through a case of customer segmentation of Chinese Petroleum Corp. HUBEI branch; the paper demonstrates that the clustering result of the case is meaningful and useful. The research is partially supported by National Natural Science Foundation of China (grants #70372049 and #70121001).     

Developments in network location with mobile and congested facilities

We review four facility location problems which are motivated by urban service applications and which can be thought of as extensions of the classic Q-median problem on networks. In problems P1 and P2 it is assumed that travel times on network links change over time in a probabilistic way. In P2 it is further assumed that the facilities (servers) are movable so that they can be relocated in response to new network travel times. Problems P3 and P4 examine the Q-median problem for the case when the service capacity of the facilities is finite and, consequently, some or all of the facilities can be unavailable part of the time. In P3 the facilities have stationary home locations but in P4 they have movable locations and thus can be relocated to compensate for the unavailability of the busy facilities. We summarize our main results to date on these problems.     

ANP application for evaluating Turkish mobile communication operators

The proposed decision model, based on Analytic Network Process (ANP), has been developed in order to determine the dynamics of the Global System for Mobile Communications (GSM) market. In accordance with the sequence of ANP, first of all, the problem was structured and modeled. At the next step, a quantitative analysis was conducted with the customers in order to assess the importance of the related factors and forecast the market shares of the Turkish mobile communication operators. The forecasted market share values were compared with actual ones for the validation of the decision model. The assessed importance of the factors in a valid model would be a potential competitive advantage for the mobile operators.     

A scheme for synchronizing clocks connected by a packet communication network

Consider a communication system in which a transmitter equipment sends fixed-size packets of data at a uniform rate to a receiver equipment. Consider also that these equipments are connected by a packet-switched network, which introduces a random delay to each packet. Here we propose an adaptive clock recovery scheme able of synchronizing the frequencies and the phases of these devices, within specified limits of precision. This scheme for achieving frequency and phase synchronization is based on measurements of the packet arrival times at the receiver, which are used to control the dynamics of a digital phase-locked loop. The scheme performance is evaluated via numerical simulations performed by using realistic parameter values.     

Heuristic fuzzy-neuro network and its application to reactive navigation of a mobile robot

A novel pattern recognition approach to reactive navigation of a mobile robot is presented in this paper. A heuristic fuzzy-neuro network is developed for pattern-mapping between quantized ultrasonic sensory data and velocity commands to the robot. The design goal was to enable an autonomous mobile robot to navigate safely and efficiently to a target position in a previously unknown environment. Useful heuristic rules were combined with the fuzzy Kohonen clustering network (FKCN) to build the desired mapping between perception and motion. This method provides much faster response to unexpected events and is less sensitive to sensor misreading than conventional approaches. It allows continuous, fast motion of the mobile robot without any need to stop for obstacles. The effectiveness of the proposed method is demonstrated in a series of practical tests on our experimental mobile robot.     

Outlier detection in two-stage semiparametric DEA models

In the use of peer group data to assess individual, typical or best practice performance, the effective detection of outliers is critical for achieving useful results, particularly for two-stage analyses. In the DEA-related literature, prior work on this issue has focused on the efficient frontier as a basis for detecting outliers. An iterative approach for dealing with the potential for one outlier to mask the presence of another has been proposed but not demonstrated. This paper proposes using both the efficient frontier and the inefficient frontier to identify outliers and thereby improve the accuracy of second stage results in two-stage nonparametric analysis. The iterative outlier detection approach is implemented in a leave-one-out method using both the efficient frontier and the inefficient frontier and demonstrated in a two-stage semi-parametric bootstrapping analysis of a classic data set. The results show that the conclusions drawn can be different when outlier identification includes consideration of the inefficient frontier.     

对恐怖袭击事件记录数据的量化分析与研究

恐怖主义是人类的共同威胁,利用数据挖掘可以为反恐防恐提供有价值的信息支持.基于数据挖掘的思路,从恐怖袭击事件中提取能描述危险程度的特征属性,构建量化分级模型,并考虑准确率评价指标进行优化.通过组内平方和法分析改进高斯混合模型(GMM),对恐怖组织进行聚类分析,侦查出潜在最相关的嫌疑人.建立相关模型结合统计分析,得到恐怖袭击发生的主要原因、时空特性和蔓延特性,并对未来全球反恐态势进行预测,帮助反恐组织提高反恐的精准性和打击能力.     

Forward search outlier detection in data envelopment analysis

In this paper we tackle the problem of outlier detection in data envelopment analysis (DEA). We propose a procedure where we merge the super-efficiency DEA and the forward search. Since DEA provides efficiency scores which are not parameters to fit the model to the data, we introduce a distance, to be monitored along the search. This distance is obtained through the integration of a regression model and the super-efficiency DEA. We simulate a Cobb-Douglas production function and we compare the super-efficiency DEA and the forward search analysis in both uncontaminated and contaminated settings. For inference about outliers, we exploit envelopes obtained through Monte Carlo simulations.     

Accelerated algorithm for pattern detection in logical analysis of data

Sets of “positive” and “negative” points (observations) in n-dimensional discrete space given along with their non-negative integer multiplicities are analyzed from the perspective of the Logical Analysis of Data (LAD). A set of observations satisfying upper and/or lower bounds imposed on certain components is called a positive pattern if it contains some positive observations and no negative one. The number of variables on which such restrictions are imposed is called the degree of the pattern. A total polynomial algorithm is proposed for the enumeration of all patterns of limited degree, and special efficient variants of it for the enumeration of all patterns with certain “sign” and “coverage” requirements are presented and evaluated on a publicly available collection of benchmark datasets.     

Metabolic Systemic Computing: Exploiting Innate Immunity within an Artificial Organism for On-line Self-Organisation and Anomaly Detection

Previous work suggests that innate immunity and representations of tissue can be useful when combined with artificial immune systems. Here we provide a new implementation of tissue for artificial immune systems using systemic computation, a new model of computation and corresponding computer architecture based on a systemics world-view and supplemented by the incorporation of natural characteristics. We show using systemic computation how to create an artificial organism, a program with metabolism that eats data, expels waste, self-organise cells based on the nature of its food and emits danger signals suitable for an artificial immune system. The implementation is tested by application to two standard machine learning sets and shows excellent abilities to recognise anomalies in its diet as well as a consistent datawise self-organisation.     

A decisional framework system for computer network intrusion detection

This paper presents a multi-attribute decisional framework for computer network intrusion detection. First, a cost model that allows to estimate accurately the damage resulting from a security incident is described. Then, a multi-attribute optimization algorithm is applied to select the optimal decision based on alternatives to remedy such incidents.     

Interest point detection in images using complex network analysis

This paper presents a new application of complex network theory and tools to digital image analysis and computer vision problems in order to detect interest points in digital images. We associate a weighted geometrical and fast computable complex network to each image and then we propose two different methods to locate these feature points based on both local and global (spectral) centrality measures of the corresponding network.     

Modelling and system identification of an experimental apparatus for anomaly detection in mechanical systems

This paper presents design, modelling and system identification of a laboratory test apparatus that has been constructed to experimentally validate the concepts of anomaly detection in complex mechanical systems. The test apparatus is designed to be complex in itself due to partially correlated interactions amongst its individual components and functional modules. The experiments are conducted on the test apparatus to represent operations of mechanical systems where both dynamic performance and structural durability are critical.     

Price and QoS competition in data communication services

Pricing of data communication services has been studied, but primarily in a monopolistic setting. We study the price competition in packet-switching networks with a quality-of-service (QoS) guarantee in terms of an expected per-packet delay. We propose a general framework in which service providers offering multi-class priority-based services compete to maximize their profits, while satisfying the expected delay guarantee in each class. We first examine the price competition with fixed delay guarantees and then extend it to the situation where providers compete in quality of service as well as price. In each case, we compare the duopoly pricing scheme with the case when two service providers merge to become a monopoly provider.     

Exact communication costs for consensus and leader in a tree

We continue the study of communication costs of Consensus and Leader initiated in a previous paper. We deal with all scenarios with linear complexity in a tree topology, and prove exact (as opposed to asymptotic) tight bounds for the bit and message complexities. A particular scenario depends on whether the tree size or the size parity is known to the processors.     

Optimal checkpointing interval of a communication system with rollback recovery

This paper considers a communication system which consists of many processors and studies the problem for improving its reliability by adopting the recovery techniques of checkpoint and rollback. When either processor failure or communication error has occurred, the rollback recovery for processors associated with such an event is executed to the most recent checkpoint, and so, a consistent state in the whole system is maintained. The stochastic model with the above recovery techniques is formulated, using the theory of Markov renewal processes. The mean time to take checkpoint and the expected numbers of rollback recovery caused by processor failures and communication errors are derived. Further, an optimal checkpointing interval which minimizes the expected cost is analytically discussed.     

Technical diagnostic of a fleet of vehicles using rough set theory

The paper presents a process of technical diagnostic applied to a fleet of vehicles utilized in the delivery system of express mail. It is focused on evaluation of diagnostic capacity of particular characteristics, reduction of a set of initially selected characteristics to a minimal and satisfactory subset, recognition of a technical condition of vehicles resulting in their condition-based classification. In addition, the decision rules facilitating technical diagnostic and management of a fleet of vehicles are generated and utilized. N-fold cross validation is applied to estimate the efficiency of the decision rules. The rough set theory is applied to support the diagnostic process of vehicles. Classical rough set (CRS) theory is compared with the dominance-based rough set (DRS) approach. The results of computational experiments for both approaches are compared.     

A Bayesian network model for spatial event surveillance

Methods for spatial cluster detection attempt to locate spatial subregions of some larger region where the count of some occurrences is higher than expected. Event surveillance consists of monitoring a region in order to detect emerging patterns that are indicative of some event of interest. In spatial event surveillance, we search for emerging patterns in spatial subregions.A well-known method for spatial cluster detection is Kulldorff’s [M. Kulldorff, A spatial scan statistic, Communications in Statistics: Theory and Methods 26 (6) (1997)] spatial scan statistic, which directly analyzes the counts of occurrences in the subregions. Neill et al. [D.B. Neill, A.W. Moore, G.F. Cooper, A Bayesian spatial scan statistic, Advances in Neural Information Processing Systems (NIPS) 18 (2005)] developed a Bayesian spatial scan statistic called BSS, which also directly analyzes the counts.We developed a new Bayesian-network-based spatial scan statistic, called BNetScan, which models the relationships among the events of interest and the observable events using a Bayesian network. BNetScan is an entity-based Bayesian network that models the underlying state and observable variables for each individual in a population.We compared the performance of BNetScan to Kulldorff’s spatial scan statistic and BSS using simulated outbreaks of influenza and cryptosporidiosis injected into real Emergency Department data from Allegheny County, Pennsylvania. It is an open question whether we can obtain acceptable results using a Bayesian network if the probability distributions in the network do not closely reflect reality, and thus, we examined the robustness of BNetScan relative to the probability distributions used to generate the data in the experiments. Our results indicate that BNetScan outperforms the other methods and its performance is robust relative to the probability distribution that is used to generate the data.     

On routing in VLSI design and communication networks

In this paper, we study the global routing problem in VLSI design and the multicast routing problem in communication networks. First we propose new and realistic models for both problems. In the global routing problem in VLSI design, we are given a lattice graph and subsets of the vertex set. The goal is to generate trees spanning these vertices in the subsets to minimize a linear combination of overall wirelength (edge length) and the number of bends of trees with respect to edge capacity constraints. In the multicast routing problem in communication networks, a graph is given to represent the network, together with subsets of the vertex set. We are required to find trees to span the given subsets and the overall edge length is minimized with respect to capacity constraints. Both problems are APX-hard. We present the integer linear programming (LP) formulation of both problems and solve the LP relaxations by the fast approximation algorithms for min-max resource-sharing problems in [K. Jansen, H. Zhang, Approximation algorithms for general packing problems and their application to the multicast congestion problem, Math. Programming, to appear, doi:10.1007/s10107-007-0106-8] (which is a generalization of the approximation algorithm proposed by Grigoriadis and Khachiyan [Coordination complexity of parallel price-directive decomposition, Math. Oper. Res. 2 (1996) 321-340]). For the global routing problem, we investigate the particular property of lattice graphs and propose a combinatorial technique to overcome the hardness due to the bend-dependent vertex cost. Finally, we develop asymptotic approximation algorithms for both problems with ratios depending on the best known approximation ratio for the minimum Steiner tree problem. They are the first known theoretical approximation bound results for the problems of minimizing the total costs (including both the edge and the bend costs) while spanning all given subsets of vertices.     

New approach to applying neural network in nonlinear dynamic model

In this work, radial basis function neural network (RBF-NN) is applied to emulate an extended Kalman filter (EKF) in a data assimilation scenario. The dynamical model studied here is based on the one-dimensional shallow water equation DYNAMO-1D. This code is simple when compared with an operational primitive equation models for numerical weather prediction. Although simple, the DYNAMO-1D is rich for representing some atmospheric motions, such as Rossby and gravity waves. It has been shown in the literature that the ability of the EKF to track nonlinear models depends on the frequency and accuracy of the observations and model errors. In some cases, just fourth-order moment EKF works well, but will be unwieldy when applied to high-dimensional state space. Artificial Neural Network (ANN) is an alternative solution for this computational complexity problem, once the ANN is trained offline with a high order Kalman filter, even though this Kalman filter has high computational cost (which is not a problem during ANN training phase). The results achieved in this work encourage us to apply this technique on operational model. However, it is not yet possible to assure convergence in high dimensional problems.