Elena Lazkano

Prototype Selection and Feature Subset Selection by Estimation of Distribution Algorithms. A Case Study in the Survival of Cirrhotic Patients Treated with TIPS

The Transjugular Intrahepatic Portosystemic Shunt (TIPS) is an interventional treatment for cirrhotic patients with portal hypertension. In the light of our medical staffs experience, the consequences of TIPS are not homogeneous for all the patients and a subgroup dies in the first six months after TIPS placement. An investigation for predicting the conduct of cirrhotic patients treated with TIPS is carried out using a clinical database with 107 cases and 77 attributes. We have applied a new Estimation of Distribution Algorithms based approach in order to perform a Prototype and Feature Subset Selection to improve the classification accuracy obtained using all the variables and all the cases. Used paradigms are K-Nearest Neighbours, Artificial Neural Networks and Classification Trees.

On the use of Bayesian Networks to develop behaviours for mobile robots

Bayesian Networks are models which capture uncertainties in terms of probabilities that can be used to perform reasoning under uncertainty. This paper presents an attempt to use Bayesian Networks as a learning technique to manage task execution in mobile robotics. To learn the Bayesian Network structure from data, the K2 structural learning algorithm is used, combined with three different net evaluation metrics. The experiment led to a new hybrid multiclassifying system resulting from the combination of 1-NN with the Bayesian Network, that allows one to use the power of the Bayesian Network while avoiding the computational burden of the reasoning mechanism - the so-called evidence propagation process. As an application example we present an approach of the presented paradigm to implement a door-crossing behaviour in a mobile robot using only sonar readings, in an environment with smooth walls and doors. Both the performance of the learning mechanism and the experiments run in the real robot-environment system show that Bayesian Networks are valuable learning mechanisms, able to deal with the uncertainty and variability inherent to such systems.

Learning to traverse doors using visual information

Mobile robots need to navigate in their environment in order to perform useful tasks. Doors appear in almost every office-like indoor environment and they have to be crossed often during the navigation process. We present in this paper a new approach that uses visual information to anticipate that a door has to be crossed. Combining then visual information with ultrasonic sensors, the robot approaches the door until an adequate distance is reached. Door traversing is then performed using sonar sensors. This paper describes the control architecture and the behaviors that have been implemented to obtain the door traversing behavior. Results and performance issues are explained. The experiments have been carried out with a B21 mobile robot.

Improving dynamic facial expression recognition with feature subset selection

This paper addresses the dynamic recognition of basic facial expressions in videos using feature subset selection. Feature selection has been already used by some static classifiers where the facial expression is recognized from one single image. Past work on dynamic facial expression recognition has emphasized the issues of feature extraction and classification, however, less attention has been given to the critical issue of feature selection in the dynamic scenario. The main contributions of the paper are as follows. First, we show that dynamic facial expression recognition can be casted into a classical classification problem. Second, we combine a facial dynamics extractor algorithm with a feature selection scheme for generic classifiers. We show that the paradigm of feature subset selection with a wrapper technique can improve the dynamic recognition of facial expressions. We provide evaluations of performance on real video sequences using five standard machine learning approaches: Support Vector Machines, K Nearest Neighbor, Naive Bayes, Bayesian Networks, and Classification Trees.

Histogram distance-based Bayesian Network structure learning: A supervised classification specific approach

In this work we introduce a methodology based on histogram distances for the automatic induction of Bayesian Networks (BN) from a file containing cases and variables related to a supervised classification problem. The main idea consists of learning the Bayesian Network structure for classification purposes taking into account the classification itself, by comparing the class distribution histogram distances obtained by the Bayesian Network after classifying each case. The structure is learned by applying eight different measures or metrics: the Cooper and Herskovits metric for a general Bayesian Network and seven different statistical distances between pairs of histograms. The results obtained confirm the hypothesis of the authors about the convenience of having a BN structure learning method which takes into account the existence of the special variable (the one corresponding to the class) in supervised classification problems.

Classifier Subset Selection to construct multi-classifiers by means of estimation of distribution algorithms

Abstract This paper proposes a novel approach to select the individual classifiers to take part in a Multiple-Classifier System. Individual classifier selection is a key step in the development of multi-classifiers. Several works have shown the benefits of fusing complementary classifiers. Nevertheless, the selection of the base classifiers to be used is still an open question, and different approaches have been proposed in the literature. This work is based on the selection of the appropriate single classifiers by means of an evolutionary algorithm. Different base classifiers, which have been chosen from different classifier families, are used as candidates in order to obtain variability in the classifications given. Experimental results carried out with 20 databases from the UCI Repository show how adequate the proposed approach is; Stacked Generalization multi-classifier has been selected to perform the experimental comparisons.

Classifier hierarchy learning by means of genetic algorithms

Classifier combination falls in the so called data mining area. Its aim is to combine some paradigms from the supervised classification - sometimes with a previous non-supervised data division phase - in order to improve the individual accuracy of the component classifiers. Formation of classifier hierarchies is an alternative among the several methods of classifier combination. In this paper we present a novel method to find good hierarchies of classifiers for given databases. In this new proposal, a search is performed by means of genetic algorithms, returning the best individual according to the classification accuracy over the dataset, estimated through 10-fold cross-validation. Experiments have been carried out over 14 databases from the UCI repository, showing an improvement in the performance compared to the single classifiers. Moreover, similar or better results than other approaches, such as decision tree bagging and boosting, have been obtained.

BAYES-NEAREST: A New Hybrid Classifier Combining Bayesian Network and Distance Based Algorithms

This paper presents a new hybrid classifier that combines the probability based Bayesian Network paradigm with the Nearest Neighbor distance based algorithm. The Bayesian Network structure is obtained from the data by using the K2 structural learning algorithm. The Nearest Neighbor algorithm is used in combination with the Bayesian Network in the deduction phase. For those data bases in which some variables are continuous valued, automatic discretizations of the data are performed. We show the performance of the new proposed approach compared with the Bayesian Network paradigm and with the well known Naive Bayes classifier in some standard databases; the results obtained by the new algorithm are better or equal according to the Wilcoxon statistical test.

On the use of a low-cost thermal sensor to improve Kinect people detection in a mobile robot.

Detecting people is a key capability for robots that operate in populated environments. In this paper, we have adopted a hierarchical approach that combines classifiers created using supervised learning in order to identify whether a person is in the view-scope of the robot or not. Our approach makes use of vision, depth and thermal sensors mounted on top of a mobile platform. The set of sensors is set up combining the rich data source offered by a Kinect sensor, which provides vision and depth at low cost, and a thermopile array sensor. Experimental results carried out with a mobile platform in a manufacturing shop floor and in a science museum have shown that the false positive rate achieved using any single cue is drastically reduced. The performance of our algorithm improves other well-known approaches, such as C4 and histogram of oriented gradients (HOG).

Fusing multiple image transformations and a thermal sensor with kinect to improve person detection ability

This paper proposes a novel approach to combine data from multiple low-cost sensors to detect people in a mobile robot. Robust detection of people is a key capability required for robots working in environments with people. Several works have shown the benefits of fusing data from complementary sensors. The Kinect sensor provides a rich data set at a significantly low cost, however, it has some limitations for its use on a mobile platform, mainly that people detection algorithms rely on images captured by a static camera. To cope with these limitations, this work is based on the fusion of Kinect and a thermical sensor (thermopile) mounted on top of a mobile platform. We propose the implementation of an evolutionary selection of sequences of image transformation to detect people through supervised classifiers. Experimental results carried out with a mobile platform in a manufacturing shop floor show that the percentage of wrong classified using only Kinect is drastically reduced with the classification algorithms and with the combination of the three information sources. Extra experiments are presented as well to show the benefits of the image transformation sequence idea here presented.