Pablo A. Estévez

Normalized Mutual Information Feature Selection

A filter method of feature selection based on mutual information, called normalized mutual information feature selection (NMIFS), is presented. NMIFS is an enhancement over Battitis MIFS, MIFS-U, and mRMR methods. The average normalized mutual information is proposed as a measure of redundancy among features. NMIFS outperformed MIFS, MIFS-U, and mRMR on several artificial and benchmark data sets without requiring a user-defined parameter. In addition, NMIFS is combined with a genetic algorithm to form a hybrid filter/wrapper method called GAMIFS. This includes an initialization procedure and a mutation operator based on NMIFS to speed up the convergence of the genetic algorithm. GAMIFS overcomes the limitations of incremental search algorithms that are unable to find dependencies between groups of features.

A review of feature selection methods based on mutual information

In this work, we present a review of the state of the art of information-theoretic feature selection methods. The concepts of feature relevance, redundance, and complementarity (synergy) are clearly defined, as well as Markov blanket. The problem of optimal feature selection is defined. A unifying theoretical framework is described, which can retrofit successful heuristic criteria, indicating the approximations made by each method. A number of open problems in the field are presented.

Subscription fraud prevention in telecommunications using fuzzy rules and neural networks

A system to prevent subscription fraud in fixed telecommunications with high impact on long-distance carriers is proposed. The system consists of a classification module and a prediction module. The classification module classifies subscribers according to their previous historical behavior into four different categories: subscription fraudulent, otherwise fraudulent, insolvent and normal. The prediction module allows us to identify potential fraudulent customers at the time of subscription. The classification module was implemented using fuzzy rules. It was applied to a database containing information of over 10,000 real subscribers of a major telecom company in Chile. In this database a subscription fraud prevalence of 2.2% was found. The prediction module was implemented as a multilayer perceptron neural network. It was able to identify 56.2% of the true fraudsters, screening only 3.5% of all the subscribers in the test set. This study shows the feasibility of significantly preventing subscription fraud in telecommunications by analyzing the application information and the customer antecedents at the time of application.

Online data visualization using the neural gas network

A high-quality distance preserving output representation is provided to the neural gas (NG) network. The nonlinear mapping is determined concurrently along with the codebook vectors. The adaptation rule for codebook positions in the projection space minimizes a cost function that favors the trustworthy preservation of the local topology. The proposed visualization method, called OVI-NG, is an enhancement over curvilinear component analysis (CCA). The results show that the mapping quality obtained with OVI-NG outperforms the original CCA, in terms of the trustworthiness, continuity, topographic function and topology preservation measures.

Selecting the Most Influential Nodes in Social Networks

A set covering greedy algorithm is proposed for solving the influence maximization problem in social networks. Two information diffusion models are considered: Independent Cascade Model and Linear Threshold Model. The proposed algorithm is compared with traditional maximization algorithms such as simple greedy and degree centrality using three data sets. In addition, an algorithm for mapping social networks is proposed, which allows visualizing the infection process and how the different algorithms evolve. The proposed approach is useful for mining large social networks.

Extracting Fuzzy Rules From Polysomnographic Recordings for Infant Sleep Classification

A neuro-fuzzy classifier (NFC) of sleep-wake states and stages has been developed for healthy infants of ages 6 mo and onward. The NFC takes five input patterns previously identified on 20-s epochs from polysomnographic recordings and assigns them to one out of five possible classes: Wakefulness, REM-Sleep, Non-REM Sleep Stage 1, Stage 2, and Stage 3-4. The definite criterion for a sleep state or stage to be established is duration of at least 1 min. The data set consisted of a total of 14 continuous recordings of naturally occurring naps (average duration: 143plusmn39 min), corresponding to a total of 6021 epochs. They were divided in a training, a validation and a test set with 7, 2, and 5 recordings, respectively. During supervised training, the system determined the fuzzy concepts associated to the inputs and the rules required for performing the classification, extracting knowledge from the training set, and pruning nonrelevant rules. Results on an independent test set achieved 83.9plusmn0.4% of expert agreement. The fuzzy rules obtained from the training examples without a priori information showed a high level of coincidence with the crisp rules stated by the experts, which are based on internationally accepted criteria. These results show that the NFC can be a valuable tool for implementing an automated sleep-wake classification system

Genetic design of biologically inspired receptive fields for neural pattern recognition

This paper proposes a new method for the design, through simulated evolution, of biologically inspired receptive fields in feedforward neural networks (NNs). The method is intended to enhance pattern recognition performance by creating new neural architectures specifically tuned for a particular pattern recognition problem. It proposes a combined neural architecture composed of two networks in cascade: a feature extraction network (FEN) followed by a neural classifier. The FEN is composed of several layers with receptive fields constructed by additive superposition of excitatory and inhibitory fields. A genetic algorithm (GA) is used to select receptive field parameters to improve classification performance. The parameters are receptive field size, orientation, and bias as well as the number of different receptive fields in each layer. Based on a random initial population where each individual represents a different neural architecture, the GA creates new enhanced individuals. The method is applied to handwritten digit classification and face recognition. In both problems, results show strong dependency between NN classification performance and receptive field architecture. GA selected parameters of the receptive fields produced improvements in the classification performance on the test set up to 90.8% for the problem of handwritten digit classification and up to 84.2% for the face recognition problem. On the same test sets, results were compared advantageously to standard feedforward multilayer perceptron (MLP) NNs where receptive fields are not explicitly defined. The MLP reached a maximum classification performance of 84.9% and 77.5% in both problems, respectively.

Automated Sleep-Spindle Detection in Healthy Children Polysomnograms

We present a new methodology to detect and characterize sleep spindles (SSs), based on the nonlinear algorithms, empirical-mode decomposition, and Hilbert-Huang transform, which provide adequate temporal and frequency resolutions in the electroencephalographic analysis. In addition, the application of fuzzy logic allows to emulate experts procedures. Additionally, we built a database of 56 all-night polysomnographic recordings from children for training and testing, which is among the largest annotated databases published on the subject. The database was split into training (27 recordings), validation (10 recordings), and testing (19 recordings) datasets. The SS events were marked by sleep experts using visual inspection, and these marks were used as golden standard. The overall SS detection performance on the testing dataset of continuous all-night sleep recordings was 88.2% sensitivity, 89.7% specificity, and 11.9% false-positive (FP) rate. Considering only non-REM sleep stage 2, the results showed 92.2% sensitivity, 90.1% specificity, and 8.9% FP rate. In general, our system presents enhanced results when compared with most systems found in the literature, thus improving SS detection precision significantly without the need of hypnogram information.

Face and iris localization using templates designed by particle swarm optimization

Face and iris localization is one of the most active research areas in image understanding for new applications in security and theft prevention, as well as in the development of human-machine interfaces. In the past, several methods for real-time face localization have been developed using face anthropometric templates which include face features such as eyes, eyebrows, nose and mouth. It has been shown that accuracy in face and iris localization is crucial to face recognition algorithms. An error of a few pixels in face or iris localization will produce significant reduction in face recognition rates. In this paper, we present a new method based on particle swarm optimization (PSO) to generate templates for frontal face localization in real time. The PSO templates were tested for face localization on the Yale B Face Database and compared to other methods based on anthropometric templates and Adaboost. Additionally, the PSO templates were compared in iris localization to a method using combined binary edge and intensity information in two subsets of the AR face database, and to a method based on SVM classifiers in a subset of the FERET database. Results show that the PSO templates exhibit better spatial selectivity for frontal faces resulting in a better performance in face localization and face size estimation. Correct face localization reached a rate of 97.4% on Yale B which was higher than 96.2% obtained with the anthropometric templates and much better than 60.5% obtained with the Adaboost face detection method. On the AR face subsets, different disparity errors were considered and for the smallest error, a 100% correct detection was reached in the AR-63 subset and 99.7% was obtained in the AR-564 subset. On the FERET subset a detection rate of 96.6% was achieved using the same criteria. In contrast to the Adaboost method, PSO templates were able to localize faces on high-contrast or poorly illuminated environments. Additionally, in comparison with the anthropometric templates, the PSO templates have fewer pixels, resulting in a 40% reduction in processing time thus making them more appropriate for real-time applications.

A review of learning vector quantization classifiers

In this work, we present a review of the state of the art of learning vector quantization (LVQ) classifiers. A taxonomy is proposed which integrates the most relevant LVQ approaches to date. The main concepts associated with modern LVQ approaches are defined. A comparison is made among eleven LVQ classifiers using one real-world and two artificial datasets.