Ana L. N. Fred

Combining Multiple Clusterings Using Evidence Accumulation

We explore the idea of evidence accumulation (EAC) for combining the results of multiple clusterings. First, a clustering ensemble - a set of object partitions, is produced. Given a data set (n objects or patterns in d dimensions), different ways of producing data partitions are: 1) applying different clustering algorithms and 2) applying the same clustering algorithm with different values of parameters or initializations. Further, combinations of different data representations (feature spaces) and clustering algorithms can also provide a multitude of significantly different data partitionings. We propose a simple framework for extracting a consistent clustering, given the various partitions in a clustering ensemble. According to the EAC concept, each partition is viewed as an independent evidence of data organization, individual data partitions being combined, based on a voting mechanism, to generate a new n × n similarity matrix between the n patterns. The final data partition of the n patterns is obtained by applying a hierarchical agglomerative clustering algorithm on this matrix. We have developed a theoretical framework for the analysis of the proposed clustering combination strategy and its evaluation, based on the concept of mutual information between data partitions. Stability of the results is evaluated using bootstrapping techniques. A detailed discussion of an evidence accumulation-based clustering algorithm, using a split and merge strategy based on the k-means clustering algorithm, is presented. Experimental results of the proposed method on several synthetic and real data sets are compared with other combination strategies, and with individual clustering results produced by well-known clustering algorithms.

Data clustering using evidence accumulation

We explore the idea of evidence accumulation for combining the results of multiple clusterings. Initially, n d-dimensional data is decomposed into a large number of compact clusters; the K-means algorithm performs this decomposition, with several clusterings obtained by N random initializations of the K-means. Taking the co-occurrences of pairs of patterns in the same cluster as votes for their association, the data partitions are mapped into a co-association matrix of patterns. This n/spl times/n matrix represents a new similarity measure between patterns. The final clusters are obtained by applying a MST-based clustering algorithm on this matrix. Results on both synthetic and real data show the ability of the method to identify arbitrary shaped clusters in multidimensional data.

Finding Consistent Clusters in Data Partitions

Given an arbitrary data set, to which no particular parametrical, statistical or geometrical structure can be assumed, different clustering algorithms will in general produce different data partitions. In fact, several partitions can also be obtained by using a single clustering algorithm due to dependencies on initialization or the selection of the value of some design parameter. This paper addresses the problem of finding consistent clusters in data partitions, proposing the analysis of the most common associations performed in a majority voting scheme. Combination of clustering results are performed by transforming data partitions into a co-association sample matrix, which maps coherent associations. This matrix is then used to extract the underlying consistent clusters. The proposed methodology is evaluated in the context of k-means clustering, a new clustering algorithm - voting-k-means, being presented. Examples, using both simulated and real data, show how this majority voting combination scheme simultaneously handles the problems of selecting the number of clusters, and dependency on initialization. Furthermore, resulting clusters are not constrained to be hyperspherically shaped.

Unveiling the biometric potential of finger-based ECG signals

The ECG signal has been shown to contain relevant information for human identification. Even though results validate the potential of these signals, data acquisition methods and apparatus explored so far compromise user acceptability, requiring the acquisition of ECG at the chest. In this paper, we propose a finger-based ECG biometric system, that uses signals collected at the fingers, through a minimally intrusive 1-lead ECG setup recurring to Ag/AgCl electrodes without gel as interface with the skin. The collected signal is significantly more noisy than the ECG acquired at the chest, motivating the application of feature extraction and signal processing techniques to the problem. Time domain ECG signal processing is performed, which comprises the usual steps of filtering, peak detection, heartbeat waveform segmentation, and amplitude normalization, plus an additional step of time normalization. Through a simple minimum distance criterion between the test patterns and the enrollment database, results have revealed this to be a promising technique for biometric applications.

A behavioral biometric system based on human-computer interaction

In this paper we describe a new behavioural biometric technique based on human computer interaction. We developed a system that captures the user interaction via a pointing device, and uses this behavioural information to verify the identity of an individual. Using statistical pattern recognition techniques, we developed a sequential classifier that processes user interaction, according to which the user identity is considered genuine if a predefined accuracy level is achieved, and the user is classified as an impostor otherwise. Two statistical models for the features were tested, namely Parzen density estimation and a unimodal distribution. The system was tested with different numbers of users in order to evaluate the scalability of the proposal. Experimental results show that the normal user interaction with the computer via a pointing device entails behavioural information with discriminating power, that can be explored for identity authentication.

A new cluster isolation criterion based on dissimilarity increments

This paper addresses the problem of cluster defining criteria by proposing a model-based characterization of interpattern relationships. Taking a dissimilarity matrix between patterns as the basic measure for extracting group structure, dissimilarity increments between neighboring patterns within a cluster are analyzed. Empirical evidence suggests modeling the statistical distribution of these increments by an exponential density; we propose to use this statistical model, which characterizes context, to derive a new cluster isolation criterion. The integration of this criterion in a hierarchical agglomerative clustering framework produces a partitioning of the data, while exhibiting data interrelationships in terms of a dendrogram-type graph. The analysis of the criterion is undertaken through a set of examples, showing the versatility of the method in identifying clusters with arbitrary shape and size; the number of clusters is intrinsically found without requiring ad hoc specification of design parameters nor engaging in a computationally demanding optimization procedure.

Evidence Accumulation Clustering Based on the K-Means Algorithm

The idea of evidence accumulation for the combination of multiple clusterings was recently proposed [7]. Taking the K-means as the basic algorithm for the decomposition of data into a large number, k, of compact clusters, evidence on pattern association is accumulated, by a voting mechanism, over multiple clusterings obtained by random initializations of the K-means algorithm. This produces a mapping of the clusterings into a new similarity measure between patterns. The final data partition is obtained by applying the single-link method over this similarity matrix. In this paper we further explore and extend this idea, by proposing: (a) the combination of multiple K-means clusterings using variable k; (b) using cluster lifetime as the criterion for extracting the final clusters; and (c) the adaptation of this approach to string patterns. This leads to a more robust clustering technique, with fewer design parameters than the previous approach and potential applications in a wider range of problems.

One-Lead ECG-based Personal Identification Using Ziv-Merhav Cross Parsing

The advance of falsification technology increases security concerns and gives biometrics an important role in security solutions. The electrocardiogram (ECG) is an emerging biometric that does not need liveliness verification. There is strong evidence that ECG signals contain sufficient discriminative information to allow the identification of individuals from a large population. Most approaches rely on ECG data and the fiducia of different parts of the heartbeat waveform. However non-fiducial approaches have proved recently to be also effective, and have the advantage of not relying critically on the accurate extraction of fiducia data. In this paper, we propose a new % NEW DAV non-fiducial ECG biometric identification method based on data compression techniques, namely the Ziv-Merhav cross parsing algorithm for symbol sequences (strings). Our method relies on a string similarity measure derived from algorithmic cross complexity concept and its compression-based approximation. NEW DAV We present results on real data, one-lead ECG, acquired during a concentration task, from 19 healthy individuals. Our approach achieves 100% subject recognition rate despite the existence of differentiated stress states.

Finger ECG signal for user authentication: Usability and performance

Over the past few years, the evaluation of Electrocardio-graphic (ECG) signals as a prospective biometric modality has revealed promising results. Given the vital and continuous nature of this information source, ECG signals offer several advantages to the field of biometrics; yet, several challenges currently prevent the ECG from being adopted as a biometric modality in operational settings. These arise partially due to ECG signals clinical tradition and intru-siveness, but also from the lack of evidence on the permanence of the ECG templates over time. The problem of in-trusiveness has been recently overcome with the “off-the-person” approach for capturing ECG signals. In this paper we provide an evaluation of the permanence of ECG signals collected at the fingers, with respect to the biometric authentication performance. Our experimental results on a small dataset suggest that further research is necessary to account for and understand sources of variability found in some subjects. Despite these limitations, “off-the-person” ECG appears to be a viable trait for multi-biometric or standalone biometrics, low user throughput, real-world scenarios.

Cluster Ensemble Methods: from Single Clusterings to Combined Solutions

Cluster ensemble methods attempt to find better and more robust clustering solutions by fusing information from several data partitionings. In this chapter, we address the different phases of this recent approach: from the generation of the partitions, the clustering ensemble, to the combination and validation of the combined result. While giving an overall revision of the state-of-the-art in the area, we focus on our own work on the subject. In particular, the Evidence Accumulation Clustering (EAC) paradigm is detailed and analyzed. For the validation/selection of the final partition, we focus on metrics that can quantitatively measure the consistency between partitions and combined results, and thus enabling the choice of best results without the use of additional information. Information-theoretic measures in conjunction with a variance analysis using bootstrapping are detailed and empirically evaluated. Experimental results throughout the paper illustrate the various concepts and methods addressed, using synthetic and real data and involving both vectorial and string-based data representations. We show that the clustering ensemble approach can be used in very distinct contexts with the state of the art quality results.