Person:
Nogales Moyano, Alberto

First Name

Alberto

Last Name

Nogales Moyano

Faculty

Escuela Politécnica Superior

Identifiers

Full item page

Search Results

Now showing 1 - 10 of 15

A domain categorisation of vocabularies based on a deep learning classifier.
(Journal of Information Science, 2021) Nogales Moyano, Alberto; Sicilia, Miguel Ángel; García Tejedor, Álvaro José
The publication of large amounts of open data has become a major trend nowadays. This is a consequence of pro-jects like the Linked Open Data (LOD) community, which publishes and integrates datasets using techniques like Linked Data. Linked Data publishers should follow a set of principles for dataset design. This information is described in a 2011 document that describes tasks as the consideration of reusing vocabularies. With regard to the latter, another project called Linked Open Vocabularies (LOV) attempts to compile the vocabularies used in LOD. These vocabularies have been classified by domain following the subjective criteria of LOV members, which has the inherent risk introducing personal biases. In this paper, we present an automatic classifier of vocabularies based on the main categories of the well-known knowledge source Wikipedia. For this purpose, word-embedding models were used, in combination with Deep Learning techniques. Results show that with a hybrid model of regular Deep Neural Network (DNN), Recurrent Neural Network (RNN) and Convolutional Neural Network (CNN), vocabularies could be classified with an accuracy of 93.57 per cent. Specifically, 36.25 per cent of the vocabularies belong to the Culture category.
BERT Learns From Electroencephalograms About Parkinson's Disease: Transformer-Based Models for Aid Diagnosis.
(IEEE Access, 2022) Nogales Moyano, Alberto; García Tejedor, Álvaro José; Maitín, Ana María; Pérez Morales, Antonio; Del Castillo, María Dolores; Romero Muñoz, Juan Pablo
Medicine is a complex field with highly trained specialists with extensive knowledge that continuously needs updating. Among them all, those who study the brain can perform complex tasks due to the structure of this organ. There are neurological diseases such as degenerative ones whose diagnoses are essential in very early stages. Parkinson’s disease is one of them, usually having a confirmed diagnosis when it is already very developed. Some physicians have proposed using electroencephalograms as a non-invasive method for a prompt diagnosis. The problem with these tests is that data analysis relies on the clinical eye of a very experienced professional, which entails situations that escape human perception. This research proposes the use of deep learning techniques in combination with electroencephalograms to develop a non-invasive method for Parkinson’s disease diagnosis. These models have demonstrated their good performance in managing massive amounts of data. Our main contribution is to apply models from the field of Natural Language Processing, particularly an adaptation of BERT models, for being the last milestone in the area. This model choice is due to the similarity between texts and electroencephalograms that can be processed as data sequences. Results show that the best model uses electroencephalograms of 64 channels from people without resting states and finger-tapping tasks. In terms of metrics, the model has values around 86%.
Network analysis for food safety: Quantitative and structural study of data gathered through the RASFF system in the European Union.
(Food Control, 2022) Nogales Moyano, Alberto; Mora-Cantallops, Marçal; Díaz Morón, Rodrigo; García Tejedor, Álvaro José
This paper reports a quantitative and structural analysis of data gathered on the food issues reported by the European Union members over the last forty years. The study applies statistical measures and network analysis techniques. For this purpose, a graph was constructed of how different contaminated products have been distributed through countries. The work aims to leverage insights into the structure formed by the involvement of European countries in the exchange of goods that can cause problems for populations. The results obtained show the roles of different countries in the detection of sensitive routes. In particular, the analysis identifies problematic origin countries, such as China or Turkey, whereas European countries, in general, do have good border control policies for the import/export of food.
Analyzing the Influence of Diverse Background Noises on Voice Transmission: A Deep Learning Approach to Noise Suppression.
(Applied Sciences, 2024) Nogales Moyano, Alberto; Caracuel-Cayuela, Javier; García Tejedor, Álvaro José
This paper presents an approach to enhancing the clarity and intelligibility of speech in digital communications compromised by various background noises. Utilizing deep learning techniques, specifically a Variational Autoencoder (VAE) with 2D convolutional filters, we aim to suppress background noise in audio signals. Our method focuses on four simulated environmental noise scenarios: storms, wind, traffic, and aircraft. The training dataset has been obtained from public sources (TED-LIUM 3 dataset, which includes audio recordings from the popular TED-TALK series) combined with these background noises. The audio signals were transformed into 2D power spectrograms, upon which our VAE model was trained to filter out the noise and reconstruct clean audio. Our results demonstrate that the model outperforms existing state-of-the-art solutions in noise suppression. Although differences in noise types were observed, it was challenging to definitively conclude which background noise most adversely affects speech quality. The results have been assessed with objective (mathematical metrics) and subjective (listening to a set of audios by humans) methods. Notably, wind noise showed the smallest deviation between the noisy and cleaned audio, perceived subjectively as the most improved scenario. Future work should involve refining the phase calculation of the cleaned audio and creating a more balanced dataset to minimize differences in audio quality across scenarios. Additionally, practical applications of the model in real-time streaming audio are envisaged. This research contributes significantly to the field of audio signal processing by offering a deep learning solution tailored to various noise conditions, enhancing digital communication quality.
EEGraph: An open-source Python library for modeling electroencephalograms using graphs.
(Neurocomputing, 2023) Maitin, Ana M.; Nogales Moyano, Alberto; Chazarra, Pedro; García Tejedor, Álvaro José
Background and objective Connectivity studies make it possible to identify alterations in brain connections and to associate these pathologies with different neurological disorders. However, a clinical test is necessary to obtain information about the state of the brain. Electroencephalograms (EEGs) provide this information in addition to being tests with other benefits for the patient (non-invasive, low-cost, high reproducibility). Graph theory can be used to represent both the anatomical and functional connections of the brain by means of connectivity measures. The procedure of transforming an EEG into a graph can be slightly tedious for researchers, especially when implementing different connectivity measures. Methods The open-source Python library EEGraph automatically performs the modeling of an EEG through a graph, providing its matrix and visual representation. It recognizes various EEG input formats, identifying the number of electrodes and the location of each electrode in the brain. Moreover, it allows the user to choose from 12 connectivity measures to produce the graph from the EEG, with great flexibility to define specific parameters to adapt them to each study, including EEG time-windows segmentation and separation in frequency bands. Results The EEGraph library is developed as a tool, for researchers and clinical specialists in the field of neuroscience, that provides direct information on the connectivity of the brain from electroencephalography signals. Its documentation and source code are available at https://github.com/ufvceiec/EEGRAPH. It can be installed from the Python Package Index using pip install EEGRAPH. Conclusions The EEGraph library was built aiming to facilitate the development of connectivity studies based on the modeling of electroencephalography tests through graphs. It includes a wide range of connectivity measures, which, together with the multiple output options, make EEGraph an easy to use and powerful tool with direct applications in both the clinical and neuroscience research fields.
A light method for data generation: a combination of Markov Chains and Word Embeddings.
(Procesamiento del Lenguaje Natural, 2020) Martínez García, Eva; Nogales Moyano, Alberto; Morales Escudero, Javier; García Tejedor, Álvaro José
Most of the current state-of-the-art Natural Language Processing (NLP) techniques are highly data-dependent. A significant amount of data is required for their training, and in some scenarios data is scarce. We present a hybrid method to generate new sentences for augmenting the training data. Our approach takes advantage of the combination of Markov Chains and word embeddings to produce high-quality data similar to an initial dataset. In contrast to other neural-based generative methods, it does not need a high amount of training data. Results show how our approach can generate useful data for NLP tools. In particular, we validate our approach by building Transformer-based Language Models using data from three different domains in the context of enriching general purpose chatbots.
An open-source Python library for self-organizing-maps.
(Software Impacts, 2022) García Tejedor, Álvaro José; Nogales Moyano, Alberto
Organizations have realized the importance of data analysis and its benefits. This in combination with Machine Learning algorithms has allowed us to solve problems more easily, making these processes less time-consuming. Neural networks are the Machine Learning technique that is recently obtaining very good best results. This paper describes an open-source Python library called GEMA developed to work with a type of neural network model called Self-Organizing-Maps. GEMA is freely available under GNU General Public License at GitHub (https://github.com/ufvceiec/GEMA). The library has been evaluated in different particular use cases obtaining accurate results.
A systematic review of electroencephalography open datasets and their usage with deep learning models.
(IEEE Access, 2023) Nogales Moyano, Alberto; García Tejedor, Álvaro José
Data are the main headache for machine learning, both because of their varied nature and their limited availability. The medical field brings together both situations: tables, images, text, or signals that are difficult to acquire due to the number of patients, the complexity and time of acquisition, or ethical constraints. The existence of open datasets is the best option for researchers in this field. Electroencephalograms are a good example of this situation. This paper identifies the primary open datasets of electroencephalography tests and how they are used in deep learning models. The aim is to provide structured information that can be consulted by researchers in the field (both physicians and computer scientists) to know which datasets are available, which characteristics they have, or which deep learning models could be applied to them. The process followed the PRISMA methodology for systematic reviews applying different inclusion and exclusion criteria to obtain a set of high-quality papers on which the data sets used were analyzed. The databases included in the searches were Scopus, PubMed, Web of Science (WOS), Science Direct, IEEE Explorer, and SpringerLink. In total, 37 papers were selected which included 30 datasets that have been considered. Then, the DL models used in the papers and the different characteristics of the datasets have been statistically analyzed by obtaining different measures and graphs. The most relevant conclusions are the widespread use of convolutional neural networks (the less innovative among the different models) as the main tool for EEG data analysis. Against this position, we found the use of hybrid models and the family of RNNs as techniques to use in cases of brain stimuli, classification of levels of fatigue, and diagnosis of diseases. Related to the datasets’ features, we demonstrate the difficulty in compiling this data due to the number of tests and that the minimum of channels or sampling frequency recommended to obtain good accuracies in the model should be studied.
Competencies in Higher Education: A Feature Analysis with Self-Organizing Maps.
(2020) Nogales Moyano, Alberto; García Tejedor, Álvaro José; Martín Sanz, Noemy; De Dios Alija, Teresa
Students are supposed to accomplish with a set of generic competencies when they finish their studies. One of the major challenges in Universities is to detect shortcomings in students in order to strengthen them, so they could accomplish with the competencies required for a professional career. In this paper, unsupervised machine learning techniques as Self-Organizing Maps are used to analyze features of students from the bachelor’s degree in Psychology. The approach is clusterization students’ profiles in their first course of college to identify potential improvement areas. The dataset contains 16 features from 54 individuals. Results show that clusters differentiate mostly on the organizational and social competencies on one side, and neuroticism and agreeableness on the other.
Enhancing breast cancer diagnosis with deep learning and evolutionary algorithms: A comparison of approaches using different thermographic imaging treatments.
(Multimedia Tools and Applications, 2023) Nogales Moyano, Alberto; Pérez Lara, Fernando; García Tejedor, Álvaro José
The medical field has come a long way in recent years. This fact is directly related to the application of computer science, particularly artificial intelligence. Computer vision is one of its applications with the most significant knowledge transfer to private companies or organizations. Due to the large number of tests based on images, it has multiple benefits in medical diagnosis. These benefits go from health to economics, passing through time savings. Most people know X-rays or scanners, but others have not been applied too much like thermographies. Although they are inexpensive, non-invasive, painless, and easy to implement in remote areas, their scientific evidence is not very extended. In this paper, we evaluate different approaches based on four use cases depending on which treatment we applied to the images. This step leads to various scenarios that could benefit from using advanced hybrid Artificial Intelligence models. Evaluating the solutions will not only provide us with an accurate model. Still, it will also allow us to understand further how the different thermograph information influences the diagnosis. Results show that by separating the thermography by three ranges of temperatures and using a hybrid model of convolutional neural networks and evolutive algorithms, we can achieve accuracy near 94%.