Search

This work presents an original algorithmic model of some essential features of psychogenetic theory, as was proposed by J. Piaget. Specifically, we modeled some elements of cognitive structure learning in children from birth to four months of life. We are in fact convinced that the study of well-established cognitive models of human learning can suggest new, interesting approaches to problem so far not satisfactorily solved in the field of machine learning. Further, we discussed the possible parallels between our model and subsymbolic machine learning and neuroscience. The model was implemented and tested in some simple experimental settings, with reference to the task of learning sensorimotor sequences.

The identification of social groups remains one of the main analytical themes in the analysis of social networks and, in more general terms, in the study of social organization. Traditional network approaches to group identification encounter a variety of problems when the data to be analyzed involve two-mode networks, i.e., relations between two distinct sets of objects with no reflexive relation allowed within each set. In this paper we propose a relatively novel approach to the recognition and identification of social groups in data generated by network-based processes in the context of two-mode networks. Our approach is based on a family of learning algorithms called Support Vector Machines (SVM). The analytical framework provided by SVM provides a flexible statistical environment to solve classification tasks, and to reframe regression and density estimation problems. We explore the relative merits of our approach to the analysis of social networks in the context of the well known “Southern women” (SW) data set collected by Davis Gardner and Gardner. We compare our results with those that have been produced by different analytical approaches. We show that our method, which acts as a data-independent preprocessing step, is able to reduce the complexity of the clustering problem enabling the application of simpler configurations of common algorithms.

The classification of tumoral masses and normal breast tissue is targeted. A mass detection algorithm which does not refer explicitly to shape, border, size, contrast or texture of mammographic suspicious regions is evaluated. In the present approach, classification features are embodied by the image representation used to encode suspicious regions. Classification is performed by means of a support vector machine (SVM) classifier. To investigate whether improvements can be achieved with respect to a previously proposed overcomplete wavelet image representation, a pixel and a discrete wavelet image representations are developed and tested. Evaluation is performed by extracting 6000 suspicious regions from the digital database for screening mammography (DDSM) collected by the University of South Florida (USF). More specifically, 1000 regions representing biopsy-proven tumoral masses (either benign or malignant) and 5000 regions representing normal breast tissue are extracted. Results demonstrate very high performance levels. The area Az under the receiver operating characteristic (ROC) curve reaches values of 0.973 ± 0.002, 0.948 ± 0.004 and 0.956 ± 0.003 for the pixel, discrete wavelet and overcomplete wavelet image representations, respectively. In particular, the improvement in the Az value with the pixel image representation is statistically significant compared to that obtained with the discrete wavelet and overcomplete wavelet image representations (two-tailed p-value < 0.0001). Additionally, 90% true positive fraction (TPF) values are achieved with false positive fraction (FPF) values of 6%, 11% and 7%, respectively.

In this work, we present a novel approach to mass detection in digital mammograms. The great variability of the appearance of masses is the main obstacle to building a mass detection method. It is indeed demanding to characterize all the varieties of masses with a reduced set of features. Hence, in our approach we have chosen not to extract any feature, for the detection of the region of interest; in contrast, we exploit all the information available on the image. A multiresolution overcomplete wavelet representation is performed, in order to codify the image with redundancy of information. The vectors of the very-large space obtained are then provided to a first support vector machine (SVM) classifier. The detection task is considered here as a two-class pattern recognition problem: crops are classified as suspect or not, by using this SVM classifier. False candidates are eliminated with a second cascaded SVM. To further reduce the number of false positives, an ensemble of experts is applied: the final suspect regions are achieved by using a voting strategy. The sensitivity of the presented system is nearly 80% with a false-positive rate of 1.1 marks per image, estimated on images coming from the USF DDSM database.

Using a custom light sheet fluorescence microscope, we image large stained human brain portions, labelled for NeuN and GAD67 neuronal markers, discerning the inhibitory population via neural-network based image analysis and exposing the brain connectivity.