Diego Alvarez Alvarez - Academia.edu (original) (raw)
Uploads
Papers by Diego Alvarez Alvarez
This paper presents a novel, Information Theoretic Learning (ITL) metric called Correntropy to tr... more This paper presents a novel, Information Theoretic Learning (ITL) metric called Correntropy to train an adaptive neuro-fuzzy classification architecture. Correntropy has been shown to have some relevant properties which suggest it to be used as a new interesting cost function on machine learning algorithms, leading to the Maximun Correntropy Criterion (MCC). We here pretend to investigate its application in the field of fuzzy classification, using it to perform parameter optimization of a neuro-fuzzy classifier. The resulting method was applied to medical sleep studies in order to classify the sleep state of the patient. The results show that, when using the MCC criterion, better parameterization is achieved in comparison with a Mean Squared Error (MSE)-guided cost function.
Expert Systems With Applications, 2009
The sleep apnea/hypopnea syndrome is a very common sleep disorder, characterised by disrupted bre... more The sleep apnea/hypopnea syndrome is a very common sleep disorder, characterised by disrupted breathing during sleep. Depending on the extent of the disruptions to sleep, these are classified as apneas or hypopneas. In order to locate these apneic events an analysis of respiratory signals recorded for an entire night's sleep is necessary. However, identifying and classifying apneic events is a complex task, given the error associated with the process for digitising signals, variability in expert criteria and the complexity of the signals themselves. This article describes a fuzzy-logic-based automated system for detecting apneic events and classifying them as apneas or hypopneas. The aim is to equip this system with mechanisms for dealing with imprecision and reasoning affected by uncertainty. The ultimate goal was to assist the physician in diagnosing the sleep apnea/hypopnea syndrome. Results in terms of locating events in the polysomnogram showed sensitivity and specificity of 0.87 and 0.89, respectively. A receiver operating curve index of 0.88 was obtained for the classification of events as apneas or hypopneas.
Expert Systems With Applications, 2011
Sleep studies are carried out in order to diagnose those diseases associated with the sleep. The ... more Sleep studies are carried out in order to diagnose those diseases associated with the sleep. The standard technique consists on monitoring various bio-physiological signals of the patient during sleep. The resulting recording, the polysomnography (PSG) is then analyzed offline by the physician. This supposes a very time-consuming task and therefore automation of these analyses is desirable. An arousal during sleep is defined as an abrupt shift in EEG frequency. Normal structure of sleep is altered by the presence of these events, thus being an important factor that influences on the quality of sleep. The use of computing assistance for the detection of these events on the PSG is aimed at reducing the cost of the PSG test, both in economical and human resources. In this work, a dataset containing PSGs of real patients was used for the detection of arousals in sleep. A total of 42 features were extracted from biosignals for the detection of these events. Our aim was to use different feature selection methods to eliminate the redundant features studying their influence on the identification of sleep arousals, checking whether classification could be improved. The objective is to reduce the number of features, identifying the subset of those with more relevance while preserving a good performance on the classifier. Two approximations are explored, wrappers and filters, using different methods of both, and also combinations of each of the methods by means of the union and the intersection of the relevant features obtained. The results showed that discarding the irrelevant features by these methods is feasible, reducing the dimensionality on the input space and also improving the accuracy of the classifiers.
This paper presents a novel, Information Theoretic Learning (ITL) method to model a fuzzy system ... more This paper presents a novel, Information Theoretic Learning (ITL) method to model a fuzzy system for regression tasks that minimizes the Renyi's entropy of the error signal. An architecture based on a generalization of the well-known Adaptive-Network-Based Fuzzy Inference System (ANFIS) was used to perform such a modeling. The resulting method was tested on the prediction of future values for the Mackey-Glass chaotic time series. The results show that, when using the ITL cost function, the method returns better models in comparison with a Mean Squared Error (MSE)-guided cost function.
Journal of Archaeological Science, 2009
During the last glacial maximum, the west European modern humans adapted to adverse conditions an... more During the last glacial maximum, the west European modern humans adapted to adverse conditions and their populations subsisted exploiting the natural resources available. Fish were always present as a source of animal protein, although their importance in the human diet and their contribution to the expansion of humans has not been fully explored. Based on the compilation of zooarchaeological evidences
Journal of Human Evolution, 2011
The transition from the Middle Palaeolithic (Mousterian) to the Upper Palaeolithic (Aurignacian) ... more The transition from the Middle Palaeolithic (Mousterian) to the Upper Palaeolithic (Aurignacian) has been one of the prominent themes in the archaeology of the European Palaeolithic for more than 20 years. One of the most controversial questions concerning this period is the extinction of the Neanderthals and their replacement by modern humans. In this context, Cueva del Conde, located in the northern part of the Iberian Peninsula, is an archaeo-palaeontological site that records the Mousterian to Aurignacian transition. It has been excavated since the beginning of the 20th century, first by the Conde de la Vega del Sella and systematically by a team from the University of Oviedo since 2001. Three main zones have been identified: the External Zone, dated to approximately 39 110 AE 520 BP (level N104); the Entrance Platform, dated between 38 250 AE 390 BP and 34 730 AE 500 BP; and Gallery A with a radiocarbon date of approximately 31 540 AE 400 BP (level N2a2). The small-vertebrate assemblages recovered from the waterscreening of all sediment from the excavation campaigns represent at least 21 small mammal, amphibian and squamate taxa. The small-vertebrate associations in the three zones suggest a patchy landscape, dominated by humid meadows and woodland areas with the existence of water in the vicinity of the cave. The climate shows a more continental pattern during the Mousterian, though it was milder during the Aurignacian. The small vertebrates of the Cueva del Conde Mousterian and Aurignacian levels suggest a climate that differed from modern day temperatures, between À1.1 and À4.4 C (mean annual temperature), placing these assemblages during Interstadials 9 to 7 (Is9 to 7).
Physical Review B, 1996
We propose a short-range generalization of the p-spin interaction spin-glass model. The model is ... more We propose a short-range generalization of the p-spin interaction spin-glass model. The model is well suited to test the idea that an entropy collapse is at the bottom line of the dynamical singularity encountered in structural glasses. The model is studied in three dimensions through Monte Carlo simulations, which put in evidence fragile glass behavior with stretched exponential relaxation and super-Arrhenius behavior of the relaxation time. Our data are in favor of a Vogel-Fulcher behavior of the relaxation time, related to an entropy collapse at the Kauzmann temperature. We, however, encounter difficulties analogous to those found in experimental systems when extrapolating thermodynamical data at low temperatures. We study the spin-glass susceptibility, investigating the behavior of the correlation length in the system. We find that the increase of the relaxation time is accompanied by a very slow growth of the correlation length. We discuss the scaling properties of off-equilibrium dynamics in the glassy regime, finding qualitative agreement with the mean-field theory.
This paper presents a novel, Information Theoretic Learning (ITL) metric called Correntropy to tr... more This paper presents a novel, Information Theoretic Learning (ITL) metric called Correntropy to train an adaptive neuro-fuzzy classification architecture. Correntropy has been shown to have some relevant properties which suggest it to be used as a new interesting cost function on machine learning algorithms, leading to the Maximun Correntropy Criterion (MCC). We here pretend to investigate its application in the field of fuzzy classification, using it to perform parameter optimization of a neuro-fuzzy classifier. The resulting method was applied to medical sleep studies in order to classify the sleep state of the patient. The results show that, when using the MCC criterion, better parameterization is achieved in comparison with a Mean Squared Error (MSE)-guided cost function.
Expert Systems With Applications, 2009
The sleep apnea/hypopnea syndrome is a very common sleep disorder, characterised by disrupted bre... more The sleep apnea/hypopnea syndrome is a very common sleep disorder, characterised by disrupted breathing during sleep. Depending on the extent of the disruptions to sleep, these are classified as apneas or hypopneas. In order to locate these apneic events an analysis of respiratory signals recorded for an entire night's sleep is necessary. However, identifying and classifying apneic events is a complex task, given the error associated with the process for digitising signals, variability in expert criteria and the complexity of the signals themselves. This article describes a fuzzy-logic-based automated system for detecting apneic events and classifying them as apneas or hypopneas. The aim is to equip this system with mechanisms for dealing with imprecision and reasoning affected by uncertainty. The ultimate goal was to assist the physician in diagnosing the sleep apnea/hypopnea syndrome. Results in terms of locating events in the polysomnogram showed sensitivity and specificity of 0.87 and 0.89, respectively. A receiver operating curve index of 0.88 was obtained for the classification of events as apneas or hypopneas.
Expert Systems With Applications, 2011
Sleep studies are carried out in order to diagnose those diseases associated with the sleep. The ... more Sleep studies are carried out in order to diagnose those diseases associated with the sleep. The standard technique consists on monitoring various bio-physiological signals of the patient during sleep. The resulting recording, the polysomnography (PSG) is then analyzed offline by the physician. This supposes a very time-consuming task and therefore automation of these analyses is desirable. An arousal during sleep is defined as an abrupt shift in EEG frequency. Normal structure of sleep is altered by the presence of these events, thus being an important factor that influences on the quality of sleep. The use of computing assistance for the detection of these events on the PSG is aimed at reducing the cost of the PSG test, both in economical and human resources. In this work, a dataset containing PSGs of real patients was used for the detection of arousals in sleep. A total of 42 features were extracted from biosignals for the detection of these events. Our aim was to use different feature selection methods to eliminate the redundant features studying their influence on the identification of sleep arousals, checking whether classification could be improved. The objective is to reduce the number of features, identifying the subset of those with more relevance while preserving a good performance on the classifier. Two approximations are explored, wrappers and filters, using different methods of both, and also combinations of each of the methods by means of the union and the intersection of the relevant features obtained. The results showed that discarding the irrelevant features by these methods is feasible, reducing the dimensionality on the input space and also improving the accuracy of the classifiers.
This paper presents a novel, Information Theoretic Learning (ITL) method to model a fuzzy system ... more This paper presents a novel, Information Theoretic Learning (ITL) method to model a fuzzy system for regression tasks that minimizes the Renyi's entropy of the error signal. An architecture based on a generalization of the well-known Adaptive-Network-Based Fuzzy Inference System (ANFIS) was used to perform such a modeling. The resulting method was tested on the prediction of future values for the Mackey-Glass chaotic time series. The results show that, when using the ITL cost function, the method returns better models in comparison with a Mean Squared Error (MSE)-guided cost function.
Journal of Archaeological Science, 2009
During the last glacial maximum, the west European modern humans adapted to adverse conditions an... more During the last glacial maximum, the west European modern humans adapted to adverse conditions and their populations subsisted exploiting the natural resources available. Fish were always present as a source of animal protein, although their importance in the human diet and their contribution to the expansion of humans has not been fully explored. Based on the compilation of zooarchaeological evidences
Journal of Human Evolution, 2011
The transition from the Middle Palaeolithic (Mousterian) to the Upper Palaeolithic (Aurignacian) ... more The transition from the Middle Palaeolithic (Mousterian) to the Upper Palaeolithic (Aurignacian) has been one of the prominent themes in the archaeology of the European Palaeolithic for more than 20 years. One of the most controversial questions concerning this period is the extinction of the Neanderthals and their replacement by modern humans. In this context, Cueva del Conde, located in the northern part of the Iberian Peninsula, is an archaeo-palaeontological site that records the Mousterian to Aurignacian transition. It has been excavated since the beginning of the 20th century, first by the Conde de la Vega del Sella and systematically by a team from the University of Oviedo since 2001. Three main zones have been identified: the External Zone, dated to approximately 39 110 AE 520 BP (level N104); the Entrance Platform, dated between 38 250 AE 390 BP and 34 730 AE 500 BP; and Gallery A with a radiocarbon date of approximately 31 540 AE 400 BP (level N2a2). The small-vertebrate assemblages recovered from the waterscreening of all sediment from the excavation campaigns represent at least 21 small mammal, amphibian and squamate taxa. The small-vertebrate associations in the three zones suggest a patchy landscape, dominated by humid meadows and woodland areas with the existence of water in the vicinity of the cave. The climate shows a more continental pattern during the Mousterian, though it was milder during the Aurignacian. The small vertebrates of the Cueva del Conde Mousterian and Aurignacian levels suggest a climate that differed from modern day temperatures, between À1.1 and À4.4 C (mean annual temperature), placing these assemblages during Interstadials 9 to 7 (Is9 to 7).
Physical Review B, 1996
We propose a short-range generalization of the p-spin interaction spin-glass model. The model is ... more We propose a short-range generalization of the p-spin interaction spin-glass model. The model is well suited to test the idea that an entropy collapse is at the bottom line of the dynamical singularity encountered in structural glasses. The model is studied in three dimensions through Monte Carlo simulations, which put in evidence fragile glass behavior with stretched exponential relaxation and super-Arrhenius behavior of the relaxation time. Our data are in favor of a Vogel-Fulcher behavior of the relaxation time, related to an entropy collapse at the Kauzmann temperature. We, however, encounter difficulties analogous to those found in experimental systems when extrapolating thermodynamical data at low temperatures. We study the spin-glass susceptibility, investigating the behavior of the correlation length in the system. We find that the increase of the relaxation time is accompanied by a very slow growth of the correlation length. We discuss the scaling properties of off-equilibrium dynamics in the glassy regime, finding qualitative agreement with the mean-field theory.