DTSC - GPM - Capítulos de Monografías
http://hdl.handle.net/10016/1592
2014-04-19T11:58:51ZGene Expressio Array Exploration Using K-Formal Concept Analysis
http://hdl.handle.net/10016/13252
Gene Expressio Array Exploration Using K-Formal Concept Analysis
González-Calabozo, José María; Peláez-Moreno, Carmen; Valverde-Albacete, Francisco J.
DNA micro-arrays are a mechanism for eliciting gene expression values, the concentration of the transcription products of a set of genes, under different chemical conditions. The phenomena of interest—up-regulation, down-regulation and co-regulation—are hypothesized to stem from the functional relationships among transcription products. In [1,2,3] a generalisation of Formal Concept Analysis was developed with data mining applications in mind, K-Formal Concept Analysis, where incidences take values in certain kinds of semirings, instead of the usual Boolean carrier set. In this paper, we use (Rmin+)- and (Rmax+) to analyse gene expression data for Arabidopsis thaliana. We introduce the mechanism to render the data in the appropriate algebra and profit by the wealth of different Galois Connections available in Generalized Formal Concept Analysis to carry different analysis for up- and down-regulated genes.
Proceeding of: 9th International Conference, ICFCA 2011, Nicosia, Cyprus, May 2-6, 2011.
2011-01-01T00:00:00ZDetecting Features from Confusion Matrices using Generalized Formal Concept Analysis
http://hdl.handle.net/10016/13110
Detecting Features from Confusion Matrices using Generalized Formal Concept Analysis
Peláez-Moreno, Carmen; Valverde-Albacete, Francisco J.
We claim that the confusion matrices of multiclass problems can be analyzed by means of a generalization of Formal Concept Analysis to obtain symbolic information about the feature sets of the underlying classification task.We prove our claims by analyzing the confusion matrices of human speech perception experiments and comparing our results to those elicited by experts.
2010-01-01T00:00:00ZSpectral lattices of (R)over-bar(max),(+)-Formal contexts
http://hdl.handle.net/10016/8787
Spectral lattices of (R)over-bar(max),(+)-Formal contexts
Valverde-Albacete, Francisco J.; Peláez-Moreno, Carmen
In [13] a generalisation of Formal Concept Analysis was introduced with data mining applications in mind, K-Formal Concept Analysis, where incidences take values in certain kinds of semirings, instead of the standard Boolean carrier set. Subsequently, the structural lattice of such generalised contexts was introduced in [15], to provide a limited equivalent to the main theorem of K-Formal Concept Analysis, resting on a crucial parameter, the degree of existence of the object-attribute pairs phi. In this paper we introduce the spectral lattice of a concrete instance of K-Formal Concept Analysis, as a further means to clarify the structural and the K-Concept Lattices and the choice of p. Specifically, we develop techniques to obtain the join- and meet-irreducibles of a (R) over bar (max),+-Concept Lattice independently of phi and try to clarify its relation to the corresponding structural lattice.
Proceeding of: 6th International Conference, ICFCA 2008, Montreal, Canada, February 25-28, 2008
2008-02-15T00:00:00ZSVMs for Automatic Speech Recognition: a Survey
http://hdl.handle.net/10016/2360
SVMs for Automatic Speech Recognition: a Survey
Solera Ureña, R.; Padrell Sendra, J.; Martín Iglesias, D.; Gallardo-Antolín, Asunción; Peláez-Moreno, Carmen; Días de María, F.
Hidden Markov Models (HMMs) are, undoubtedly, the most employed core technique for Automatic Speech Recognition (ASR). Nevertheless, we are still far from achieving high-performance ASR systems. Some alternative approaches, most of them based on Artificial Neural Networks (ANNs), were proposed during the late eighties and early nineties. Some of them tackled the ASR problem using predictive ANNs, while others proposed hybrid HMM/ANN systems. However, despite some achievements, nowadays, the preponderance of Markov Models is a fact.
During the last decade, however, a new tool appeared in the field of machine learning that has proved to be able to cope with hard classification problems in several fields of application: the Support Vector Machines (SVMs). The SVMs are effective discriminative classifiers with several outstanding characteristics, namely: their solution is that with maximum margin; they are capable to deal with samples of a very higher dimensionality; and their convergence to the minimum of the associated cost function is guaranteed.
These characteristics have made SVMs very popular and successful. In this chapter we discuss their strengths and weakness in the ASR context and make a review of the current state-of-the-art techniques. We organize the contributions in two parts: isolated-word recognition and continuous speech recognition. Within the first part we review several techniques to produce the fixed-dimension vectors needed for original SVMs. Afterwards we explore more sophisticated techniques based on the use of kernels capable to deal with sequences of different length. Among them is the DTAK kernel, simple and effective, which rescues an old technique of speech recognition: Dynamic Time Warping (DTW). Within the second part, we describe some recent approaches to tackle more complex tasks like connected digit recognition or continuous speech recognition using SVMs. Finally we draw some conclusions and outline several ongoing lines of research.
2007-01-01T00:00:00Z