DigitalUM :: Browsing by Subject "Deep learning"

Browsing by Subject "Deep learning"

Now showing 1 - 20 of 21

Open Access
A self-adaptive deep learning-based system for anomaly detection in 5G networks
(IEEE, 2018-02-07) Fernández Maimó, Lorenzo; Perales Gómez, Ángel Luis; García Clemente, Félix J.; Gil Pérez, Manuel; Martínez Pérez, Gregorio; Ingeniería y Tecnología de Computadores
The upcoming fifth-generation (5G) mobile technology, which includes advanced communication features, is posing new challenges on cybersecurity defense systems. Although innovative approaches have evolved in the last few years, 5G will make existing intrusion detection and defense procedures become obsolete, in case they are not adapted accordingly. In this sense, this paper proposes a novel 5G-oriented cyberdefense architecture to identify cyberthreats in 5G mobile networks efficient and quickly enough. For this, our architecture uses deep learning techniques to analyze network traffic by extracting features from network flows. Moreover, our proposal allows adapting, automatically, the configuration of the cyberdefense architecture in order to manage traffic fluctuation, aiming both to optimize the computing resources needed in each particular moment and to fine tune the behavior and the performance of analysis and detection processes. Experiments using a well-known botnet data set depict how a neural network model reaches a sufficient classification accuracy in our anomaly detection system. Extended experiments using diverse deep learning solutions analyze and determine their suitability and performance for different network traffic loads. The experimental results show how our architecture can self-adapt the anomaly detection system based on the volume of network flows gathered from 5G subscribers' user equipments in real-time and optimizing the resource consumption.
Open Access
A time series forecasting based multi-criteria methodology for air quality prediction
(Elsevier, 2021-09-07) Espinosa Fernández, Raquel; Palma Méndez, José Tomás; Jiménez Barrionuevo, Fernando; Kamińska, Joanna; Sciavicco, Guido; Lucena Sánchez, Estrella; Ingeniería de la Información y las Comunicaciones
There is a very extensive literature on the design and test of models of environmental pollution, especially in the atmosphere. Current and recent models, however, are focused on explaining the causes and their temporal relationships, but do not explore, in full detail, the performances of pure forecasting models. We consider here three years of data that contain hourly nitrogen oxides concentrations in the air; exposure to high concentrations of these pollutants has been indicated as potential cause of numerous respiratory, circulatory, and even nervous diseases. Nitrogen oxides concentrations are paired with meteorological and vehicle traffic data for each measure. We propose a methodology based on exactness and robustness criteria to compare different pollutant forecasting models and their characteristics. 1DCNN, GRU and LSTM deep learning models, along with Random Forest, Lasso Regression and Support Vector Machines regression models, are analyzed with different window sizes. As a result, our best models offer a 24-hours ahead, very reliable prediction of the concentration of pollutants in the air in the considered area, which can be used to plan, and implement, different kinds of interventions and measures to mitigate the effects on the population.
Open Access
An interpretable semi‐supervised system for detecting cyberattacks using anomaly detection in industrial scenarios
(Wiley Open Access, 2023-05-09) Perales Gómez, Ángel Luis; Fernández Maimó, Lorenzo; García Clemente, Félix J.; Huertas Celdrán, Alberto; Ingeniería y Tecnología de Computadores
When detecting cyberattacks in Industrial settings, it is not sufficient to determine whether the system is suffering a cyberattack. It is also fundamental to explain why the system is under a cyberattack and which are the assets affected. In this context, the Anomaly Detection based on Machine Learning (ML) and Deep Learning (DL) techniques showed great performance when detecting cyberattacks in industrial scenarios. However, two main limitations hinder using them in a real environment. Firstly, most solutions are trained using a supervised approach, which is impractical in the real industrial world. Secondly, the use of black-box ML and DL techniques makes it impossible to interpret the decision made by the model. This article proposes an interpretable and semi-supervised system to detect cyberattacks in Industrial settings. Besides, our proposal was validated using data collected from the Tennessee Eastman Process. To the best of our knowledge, this system is the only one that offers interpretability together with a semi-supervised approach in an industrial setting. Our system discriminates between causes and effects of anomalies and also achieved the best performance for 11 types of anomalies out of 20 with an overall recall of 0.9577, a precision of 0.9977, and a F1-score of 0.9711.
Restricted
AuthCODE: a privacy-preserving and multi-device continuous authentication architecture based on machine and deep learning
(Elsevier, 2021-01-04) Sánchez Sánchez, Pedro Miguel; Fernández Maimó, Lorenzo; Martínez Pérez, Gregorio; Huertas Celdrán, Alberto; Ingeniería y Tecnología de Computadores
The authentication field is evolving towards mechanisms able to keep users continuously authenticated without the necessity of remembering or possessing authentication credentials. While relevant limitations of continuous authentication systems -high false positives rates (FPR) and difficulty to detect behaviour changes- have been demonstrated in realistic single-device scenarios, the Internet of Things and next generation of mobile networks (5G) are enabling novel multi-device scenarios, such as Smart Offices, that can help to reduce or address the previous challenges. The paper at hand presents an AI-based, privacy-preserving and multi-device continuous authentication architecture called AuthCODE. AuthCODE seeks to improve single-device solutions limitations by considering additional behavioural data coming from heterogeneous devices. AuthCODE proposes a novel set of features that combine the interactions of users with different devices. The features relevance has been demonstrated in a realistic Smart Office scenario with several users that interact with their mobile devices and personal computers. In this context, a set of single- and multi-device datasets have been generated and published to compare the performance of our multi-device solution against single-device approaches. A pool of experiments with machine and deep learning classifiers measured the impact of time in authentication accuracy and improved the results of single-device approaches by considering multi-device behaviour profiles. Specifically, the multi-device approach using XGBoost with 1-minute window of aggregated features, achieved a 69.33%, 59,65% and 89,35% improvement in the FPR when compared to the single-device approach for computer, mobile applications and mobile sensors respectively. Finally, temporal information classified by a Long-Short Term Memory Network, allowed the identification of additional complex behaviour patterns.
Open Access
Comparison of 2D and 3D convolutional neural networks in hyperspectral image analysis of fruits applied to orange bruise detection
(Wiley, 2023-10-25) Pourdarbani, Raziyeh; Sabzi, Sajad; Zohrabi, Reihaneh; García Mateos, Ginés; Fernández Beltrán, Rubén; Molina Martínez, José Miguel; Rohban, Mohammad H.; Informática y Sistemas; Facultades de la UMU::Facultad de Informática
Recent advances in hyperspectral imaging (HSI) have demonstrated its ability to detect defects in fruit that may not be visible in RGB images. HSIs can be considered 3D images containing two spatial dimensions and one spectral dimension. Therefore, the first question that arises is how to process this type of information, either using 2D or 3D models. In this study, HSI in the 550–900 nm spectral range was used to detect bruising in oranges. Sixty samples of Thompson oranges were subjected to a mechanical bruising process, and HSIs were taken at different time intervals: before bruising, and 8 and 16 h after bruising. The samples were then classified using two convolutional neural network (CNN) models, a shallow 7-layer network (CNN-7) and a deep 18-layer network (CNN-18). In addition, two different input processing approaches are used: using 2D information from each band, and using the full 3D data from each HSI. The 3D models were the most accurate, with 94% correct classification for 3D-CNN-18, compared to 90% for 3D-CNN-7, and less than 83% for the 2D models. Our study suggests that 3D HSI may be a more effective technique for detecting fruit bruising, allowing the development of a fast, accurate, and nondestructive method for fruit sorting.
Open Access
Convolutional neural networks for estimating the ripening state of fuji apples using visible and near-infrared spectroscopy
(Springer , 2022-07-18) Benmouna, Brahim; García Mateos, Ginés; Sabzi, Sajad; Fernández Beltrán, Rubén; Parras Burgos, Dolores; Molina Martínez, José Miguel; Informática y Sistemas; Facultades de la UMU::Facultad de Informática
The quality of fresh apple fruits is a major concern for consumers and manufacturers. Classification of these fruits according to their ripening stage is one of the most decisive factors in determining their quality. In this regard, the aim of this work is to develop a new method for non-destructive classification of the ripening state of Fuji apples using hyperspectral information in the visible and near-infrared (Vis/NIR) regions. Spectra of 172 apple samples in the range from 450 to 1000 nm were studied, which were selected from four different ripening stages. A convolutional neural network (CNN) model was proposed to perform the classification of the samples. The proposed method was compared with three alternative methods based on artificial neural networks (ANN), support vector machines (SVM), and k-nearest neighbors (KNN). The results revealed that the CNN method outperformed the alternative methods, achieving a correct classification rate (CCR) of 96.5%, compared with an average of 89.5%, 95.93%, and 91.68% for ANN, SVM, and KNN, respectively. These results will help in the development of a new device for fast and accurate estimation of the quality of apples.
Open Access
Deep learning-based postoperative glioblastoma segmentation and extent of resection evaluation: development, external validation, and model comparison
(Oxford University Press, 2024-11-16) Cepeda, Santiago; Romero, Roberto; Luque, Lidia; García Pérez, Daniel; Blasco, Guillermo; Tommaso Luppino, Luigi; Kuttner, Samuel; Estéban-Sinovas, Olga; Arrese, Ignacio; Solheim, Ole; Eikenes, Live; Karlberg, Anna; Pérez-Núñez, Ángel; Zanier, Olivier; Serra, Carlo; Staartjes, Victor E.; Bianconi, Andrea; Rossi, Luca Francesco; Garbossa, Diego; Escudero, Trinidad; Hornero, Roberto; Sarabia, Rosario; Farmacología; Farmacia
Background: The pursuit of automated methods to assess the extent of resection (EOR) in glioblastomas is challenging, requiring precise measurement of residual tumor volume. Many algorithms focus on preoperative scans, making them unsuitable for postoperative studies. Our objective was to develop a deep learning-based model for postoperative segmentation using magnetic resonance imaging (MRI). We also compared our model’s performance with other available algorithms. Methods: To develop the segmentation model, a training cohort from 3 research institutions and 3 public databases was used. Multiparametric MRI scans with ground truth labels for contrast-enhancing tumor (ET), edema, and surgical cavity, served as training data. The models were trained using MONAI and nnU-Net frameworks. Comparisons were made with currently available segmentation models using an external cohort from a research institution and a public database. Additionally, the model’s ability to classify EOR was evaluated using the RANO-Resect classification system. To further validate our best-trained model, an additional independent cohort was used. Results: The study included 586 scans: 395 for model training, 52 for model comparison, and 139 scans for independent validation. The nnU-Net framework produced the best model with median Dice scores of 0.81 for contrast ET, 0.77 for edema, and 0.81 for surgical cavities. Our best-trained model classified patients into maximal and submaximal resection categories with 96% accuracy in the model comparison dataset and 84% in the independent validation cohort. Conclusions: Our nnU-Net-based model outperformed other algorithms in both segmentation and EOR classification tasks, providing a freely accessible tool with promising clinical applicability.
Restricted
Dynamic management of a deep learning-based anomaly detection system for 5G networks
(Springer, 2018-05-05) Fernández Maimó, Lorenzo; Gil Pérez, Manuel; García Clemente, Félix Jesús; Martínez Pérez, Gregorio; Huertas Celdrán, Alberto; Ingeniería y Tecnología de Computadores
Fog and mobile edge computing (MEC) will play a key role in the upcoming fifth generation (5G) mobile networks to support decentralized applications, data analytics and management into the network itself by using a highly distributed compute model. Furthermore, increasing attention is paid to providing user-centric cybersecurity solutions, which particularly require collecting, processing and analyzing significantly large amount of data traffic and huge number of network connections in 5G networks. In this regard, this paper proposes a MEC-oriented solution in 5G mobile networks to detect network anomalies in real-time and in autonomic way. Our proposal uses deep learning techniques to analyze network flows and to detect network anomalies. Moreover, it uses policies in order to provide an efficient and dynamic management system of the computing resources used in the anomaly detection process. The paper presents relevant aspects of the deployment of the proposal and experimental results to show its performance.
Open Access
FARMIT: Continuous Assessment of Crop Quality Using Machine Learning and Deep Learning Techniques for IoT-based Smart Farming
(Springer, 2022-03-31) Perales Gómez, Ángel Luis; López de Teruel Alcolea, Pedro Enrique; Ruiz García, Alberto; García Mateos, Ginés; García Clemente, Félix Jesús; Ingeniería y Tecnología de Computadores
The race for automation has reached farms and agricultural fields. Many of these facilities use the Internet of Things (IoT) technologies to automate processes and increase productivity. Besides, Machine Learning and Deep Learning allow performing continuous decision making based on data analysis. In this work, we fill a gap in the literature and present a novel architecture based on IoT and Machine Learning / Deep Learning technologies or the continuous assessment of agricultural crop quality. This architecture is divided into three layers that work together to gather, process, and analyze data from different sources to evaluate crop quality. In the experiments, he proposed approach based on data aggregation from different sources reaches a lower percentage error than considering only one source. In particular, the percentage error achieved by our approach in the test dataset was 6.59, while the percentage error achieved exclusively using data from sensors was 6.71.
Open Access
Federated vs local vs central deep learning of tooth segmentation on panoramic radiographs
(Elsevier, 2023-05-18) Schneider, Lisa; Rischke, Roman; Krois, Joachim; Krasowki, Aleksander; Büttmer, Martha; Mohammad-Rahimi, Hossein; Chaurasai, Akhilanand; Pereira, Nielsen, S.; Lee, Jae-hong; Uribe, Sergio E.; Shahab, Shahriar; Birke Koca-Ünsal, Revan; Ünsal, Gürkan; Martínez Beneyto, Yolanda; Brinz, Janet; Tryfonos, Olga; Schwendicke, Falk; Dermatología, Estomatología, Radiología y Medicina Física
Objective: Federated Learning (FL) enables collaborative training of artificial intelligence (AI) models from multiple data sources without directly sharing data. Due to the large amount of sensitive data in dentistry, FL may be particularly relevant for oral and dental research and applications. This study, for the first time, employed FL for a dental task, automated tooth segmentation on panoramic radiographs. Methods: We employed a dataset of 4,177 panoramic radiographs collected from nine different centers (n = 143 to n = 1881 per center) across the globe and used FL to train a machine learning model for tooth segmentation. FL performance was compared against Local Learning (LL), i.e., training models on isolated data from each center (assuming data sharing not to be an option). Further, the performance gap to Central Learning (CL), i.e., training on centrally pooled data (based on data sharing agreements) was quantified. Generalizability of models was evaluated on a pooled test dataset from all centers. Results: For 8 out of 9 centers, FL outperformed LL with statistical significance (p<0.05); only the center providing the largest amount of data FL did not have such an advantage. For generalizability, FL outperformed LL across all centers. CL surpassed both FL and LL for performance and generalizability. Conclusion: If data pooling (for CL) is not feasible, FL is shown to be a useful alternative to train performant and, more importantly, generalizable deep learning models in dentistry, where data protection barriers are high. Clinical Significance: This study proves the validity and utility of FL in the field of dentistry, which encourages researchers to adopt this method to improve the generalizability of dental AI models and ease their transition to the clinical environment.
Open Access
Fedstellar: a platform for decentralized federated learning
(Elsevier, 2024-05-14) Martínez Beltrán, Enrique Tomás; Perales Gómez, Ángel Luis; Feng, Chao; Sánchez Sánchez, Pedro Miguel; López Bernal, Sergio; Gérôme, Bovet; Gil Pérez, Manuel; Martínez Pérez, Gregorio; Huertas Celdrán, Alberto; Ingeniería y Tecnología de Computadores
In 2016, Google proposed Federated Learning (FL) as a novel paradigm to train Machine Learning (ML) models across the participants of a federation while preserving data privacy. Since its birth, Centralized FL (CFL) has been the most used approach, where a central entity aggregates participants’ models to create a global one. However, CFL presents limitations such as communication bottlenecks, single point of failure, and reliance on a central server. Decentralized Federated Learning (DFL) addresses these issues by enabling decentralized model aggregation and minimizing dependency on a central entity. Despite these advances, current platforms training DFL models struggle with key issues such as managing heterogeneous federation network topologies, adapting the FL process to virtualized or physical deployments, and using a limited number of metrics to evaluate different federation scenarios for efficient implementation. To overcome these challenges, this paper presents Fedstellar, a novel platform designed to train FL models in a decentralized, semi-decentralized, and centralized fashion across diverse federations of physical or virtualized devices. Fedstellar allows users to create federations by customizing parameters like the number and type of devices training FL models, the network topology connecting them, the machine and deep learning algorithms, or the datasets of each participant, among others. Additionally, it offers real-time monitoring of model and network performance. The Fedstellar implementation encompasses a web application with an interactive graphical interface, a controller for deploying federations of nodes using physical or virtual devices, and a core deployed on each device, which provides the logic needed to train, aggregate, and communicate in the network. The effectiveness of the platform has been demonstrated in two scenarios: a physical deployment involving single-board devices such as Raspberry Pis for detecting cyberattacks and a virtualized deployment comparing various FL approaches in a controlled environment using MNIST and CIFAR-10 datasets. In both scenarios, Fedstellar demonstrated consistent performance and adaptability, achieving of 91%, 98%, and 91.2% using DFL for detecting cyberattacks and classifying MNIST and CIFAR-10, respectively, reducing training time by 32% compared to centralized approaches.
Open Access
Hope speech detection in Spanish. The LGBT case
(Springer, 2023-03-17) García‑Baena, Daniel; García‑Cumbreras, Miguel Ángel; Jiménez‑Zafra, Salud María; García Díaz, José Antonio; Valencia García, Rafael; Informática y Sistemas; Facultad de Informática
In recent years, systems have been developed to monitor online content and remove abusive, offensive or hateful content. Comments in online social media have been analyzed to find and stop the spread of negativity using methods such as hate speech detection, identification of offensive language or detection of abusive language. We define hope speech as the type of speech that is able to relax a hostile environment and that helps, gives suggestions and inspires for good to a number of people when they are in times of illness, stress, loneliness or depression. Detecting it automatically, in order to give greater diffusion to positive comments, can have a very significant effect when it comes to fighting against sexual or racial discrimination or when we intend to foster less bellicose environments. In this article we perform a complete study on hope speech in Spanish, analyzing existing solutions and available resources. In addition, we have generated a quality resource, a new Twitter dataset on LGBT community, and we have conducted some experiments that can serve as a baseline for further research.
Open Access
Ontology-driven aspect-based sentiment analysis classification: an infodemiological case study regarding infectious diseases in Latin America
(Elsevier, 2020-06-14) García Díaz, José Antonio; Cánovas García, Mar; Valencia García, Rafael; Informática y Sistemas; Facultades de la UMU::Facultad de Informática
Infodemiology is the process of mining unstructured and textual data so as to provide public health officials and policymakers with valuable information regarding public health. The appearance of this new data source, which was previously unimaginable, has opened up a new way in which to improve public health systems, resulting in better communication policies and better detection systems. However, the unstructured nature of the Internet, along with the complexity of the infectious disease domain, prevents the information extracted from being easily understood. Moreover, when dealing with languages other than English, for which some of the most common Natural Language Processing resources are not available, the correct exploitation of this data becomes even more difficult. We intend to fill these gaps proposing an ontology-driven aspect-based sentiment analysis with which to measure the general public’s opinions as regards infectious diseases when expressed in Spanish by employing a case study of tweets concerning the Zika, Dengue and Chikungunya viruses in Latin America. Our proposal is based on two technologies. We first use ontologies in order to model the infectious disease domain with concepts such as risks, symptoms, transmission methods or drugs, among other concepts. We then measure the relationship between these concepts in order to determine the degree to which one concept influences other concepts. This new information is subsequently applied in order to build an aspect-based sentiment analysis model based on statistical and linguistic features. This is done by applying deep-learning models. Our proposal is available on a web platform, where users can see the sentiment for each concept at a glance and analyse how each concept influences the sentiment of the others.
Open Access
SafeMan: a unified framework to manage cybersecurity and safety in manufacturing industry
(Wiley, 2020-08-06) Perales Gómez, Ángel Luis; Fernández Maimó, Lorenzo; García Clemente, Félix Jesús; Gil Pérez, Manuel; Martínez Pérez, Gregorio; Huertas Celdrán, Alberto; Ingeniería y Tecnología de Computadores
Industrial control systems (ICS) are considered cyber-physical systems that join both cyber and physical worlds. Due to their tight interaction, where humans and robots co-work and co-inhabit in the same workspaces and production lines, cyber-attacks targeting ICS can alter production processes and even bypass safety procedures. As an example, these cyber-attacks could interrupt physical industrial processes and cause potential injuries to workers. In this article, we present SafeMan, a unified management framework based on the Edge Computing paradigm that provides high-performance applications for the detection and mitigation of both cyber-attacks and safety threats in industrial scenarios. Three use cases show specific threats in manufacturing as well as the SafeMan actions carried out to detect and mitigate them. In order to validate our proposal, a pool of experiments was performed with Electra, an industrial dataset with normal network traffic and different cyber-attacks by using a given number of Modbus TCP and S7Comm devices. The experiments measured the runtime performance of anomaly detection techniques based on machine learning and deep learning to detect cyber-attacks in control networks. The experimental results show that Neural Networks report the best performance, being able to examine 217 feature vectors per second over Electra, and therefore demonstrating that it can be used as detection model for SafeMan in real scenarios.
Open Access
Simulating the cellular context in synthetic datasets for cryo-electron tomography
(Institute of Electrical and Electronics Engineers, 2024-05-08) Martinez-Sanchez, Antonio; Lamm, Lorenz; Jasnin, Marion; Phelippeau, Harold; Ingeniería de la Información y las Comunicaciones
Cryo-electron tomography (cryo-ET) allows to visualize the cellular context at macromolecular level. To date, the impossibility of obtaining a reliable ground truth is limiting the application of deep learning-based image processing algorithms in this field. As a consequence, there is a growing demand of realistic synthetic datasets for training deep learning algorithms. In addition, besides assisting the acquisition and interpretation of experimental data, synthetic tomograms are used as reference models for cellular organization analysis from cellular tomograms. Current simulators in cryo-ET focus on reproducing distortions from image acquisition and tomogram reconstruction, however, they can not generate many of the low order features present in cellular tomograms. Here we propose several geometric and organization models to simulate low order cellular structures imaged by cryo-ET. Specifically, clusters of any known cytosolic or membrane bound macromolecules, membranes with different geometries as well as different filamentous structures such as microtubules or actin-like networks. Moreover, we use parametrizable stochastic models to generate a high diversity of geometries and organizations to simulate representative and generalized datasets, including very crowded environments like those observed in native cells. These models have been implemented in a multiplatform open-source Python package, including scripts to generate cryo-tomograms with adjustable sizes and resolutions. In addition, these scripts provide also distortion-free density maps besides the ground truth in different file formats for efficient access and advanced visualization. We show that such a realistic synthetic dataset can be readily used to train generalizable deep learning algorithms.
Open Access
Smart analysis of economics sentiment in Spanish based on linguistic features and transformers
(IEEE, 2023-02-10) García Díaz, José Antonio; García-Sánchez, Francisco ; Valencia García, Rafael; Informática y Sistemas; Facultad de Informática
Texts related to economics and finances are characterized by the use of words and expressions whose meaning (and the sentiments they convey) substantially depend on the context. This poses a major challenge to Natural Language Processing tasks in general, and Sentiment Analysis in particular. For lowresource languages such as Spanish, this situation becomes even more acute. Yet, the latest advancements in the field, including word embeddings and transformers, have allowed to boost the performance of Sentiment Analysis solutions. In this work we explore the impact of the combination of different feature sets in the accuracy of Sentiment Analysis in Spanish financial texts. For this, a corpus with 15,915 tweets has been compiled and manually annotated as either positive, negative, or neutral. Then, feature sets based on contextual and non-contextual embeddings along with linguistic features were evaluated both individually and combined. The best results, with a weighted F1-score of 73.15880%, were obtained with a combination of feature sets by means of knowledge integration
Open Access
Spanish MEACorpus 2023: a multimodal speech–text corpus for emotion analysis in Spanish from natural environments
(Elsevier, 2024-08) Pan, Ronghao; García Díaz, José Antonio; Rodríguez García, Miguel Ángel; Valencia García, Rafael; Informática y Sistemas; Facultades de la UMU::Facultad de Informática
In human–computer interaction, emotion recognition provides a deeper understanding of the user’s emotions, enabling empathetic and effective responses based on the user’s emotional state. While deep learning models have improved emotion recognition solutions, it is still an active area of research. One important limitation is that most emotion recognition systems use only text as input, ignoring features such as voice intonation. Another limitation is the limited number of datasets available for multimodal emotion recognition. In addition, most published datasets contain emotions that are simulated by professionals and produce limited results in real-world scenarios. In other languages, such as Spanish, hardly any datasets are available. Therefore, our contributions to emotion recognition are as follows. First, we compile and annotate a new corpus for multimodal emotion recognition in Spanish (Spanish MEACorpus 2023), which contains 13.16 h of speech divided into 5129 segments labeled by considering Ekman’s six basic emotions. The dataset is extracted from YouTube videos in natural environments. Second, we explore several deep learning models for emotion recognition using text- and audio-based features. Third, we evaluate different multimodal techniques to build a multimodal recognition system that improves the results of unimodal models, achieving a Macro F1-score of 87.745%, using late fusion with concatenation strategy approach.
Embargo
Surrogate-assisted multi-objective evolutionary feature selection of generation-based fixed evolution control for time series forecasting with LSTM networks
(Elsevier, 2024-05-15) Espinosa Fernández, Raquel; Jiménez Barrionuevo, Fernando; Palma Méndez, José Tomás; Ingeniería de la Información y las Comunicaciones
Surrogate-assisted multi-objective evolutionary algorithms are powerful techniques to solve computationally expensive multi-objective optimization problems. In this paper, we propose a direct fitness replacement method with generation-based fixed evolution control to implement a multi-objective evolutionary algorithm that uses a surrogate model for wrapper-type feature selection, where long short-term memory is established as the learning algorithm. The importance of the work and its benefits lie in the need to reduce the excessive computational time required by conventional wrapper-type feature selection methods based on multi-objective evolutionary algorithms and LSTM networks, maintaining or improving the predictive capacity of the models. We analyze the use of incremental learning to update the surrogate model, in comparison with the conventional non-incremental learning approach. We applied these methods in real-life time series forecasting of air quality, indoor temperature in a smart building and oil temperature in electricity transformers. Multi-step ahead predictions of the forecast models obtained with different meta-learners of the surrogate model were compared by using the Diebold–Mariano statistical test on a multi-criteria performance metric. The proposed method outperformed other approaches for feature selection including, among others, methods based on surrogate-assisted multi-objective evolutionary algorithms developed by the authors in previous research, other surrogate-assisted deterministic methods for feature selection and the conventional wrapper-type feature selection method based on LSTM, improving the prediction on test dataset by 23.98%, 34.61% and 13.77%, respectively.
Open Access
SUSAN: a deep learning based anomaly detection framework for sustainable industry
(2023-01-05) Perales Gómez, Ángel Luis; Fernández Maimó, Lorenzo; García Clemente, Félix J.; Huertas Celdrán, Alberto; Ingeniería y Tecnología de Computadores; Elsevier
Nowadays, sustainability is the core of green technologies, being a critical aspect in many industries concerned with reducing carbon emissions and energy consumption optimization. While this concern increases, the number of cyberattacks causing sustainability issues in industries also grows. These cyberattacks impact industrial systems that control and monitor the right functioning of processes and systems. Furthermore, they are very specialized, requiring knowledge about the target industrial processes, and being undetectable for traditional cybersecurity solutions. To overcome this challenge, we present SUSAN, a Deep Learning-based framework, to build anomaly detectors that expose cyberattacks affecting the sustainability of industrial systems. SUSAN follows a modular and flexible design that allows the ensembling of several detectors to achieve more precise detections. To demonstrate the feasibility of SUSAN, we implemented the framework in a water treatment plant using the SWaT testbed. The experiments performed achieved the best recall rate (0.910) and acceptable precision (0.633), resulting in an F1-score of 0.747. Regarding individual cyberattacks that impact the system’s sustainability, our implementation detected all of them, and, concerning the related work, it achieved the most balanced results, with 0.64 as the worst recall rate. Finally, a false-positive rate of 0.000388 makes our solution feasible in real scenarios.
Open Access
VAASI: Crafting valid and abnormal adversarial samples for anomaly detection systems in industrial scenarios
(Elsevier, 2023-12) Perales Gómez, Ángel Luis; Fernández Maimó, Lorenzo; García Clemente, Félix J.; Huertas Celdrán, Alberto; Ingeniería y Tecnología de Computadores
In the realm of industrial anomaly detection, machine and deep learning models face a critical vulnerability to adversarial attacks. In this context, existing attack methodologies primarily target continuous features, often in the context of images, making them unsuitable for the categorical or discrete features prevalent in industrial systems. To fortify the cybersecurity of industrial environments, this paper introduces a groundbreaking adversarial attack approach tailored to the unique demands of these settings. Our novel technique enables the creation of targeted adversarial samples that are valid within the framework of supervised cyberattack detection models in industrial scenarios, preserving the consistency of discrete values and correcting cases where an adversarial sample transitions into a normal one. Our approach leverages the SHAP interpretability method to identify the most salient features for each sample. Subsequently, the Projected Gradient Descent technique is employed to perturb continuous features, ensuring adversarial sample generation. To handle categorical features for a specific adversarial sample, our method scrutinizes the closest sample within the normal training dataset and replicates its categorical feature values. Additionally, Decision Trees trained within a Random Forest are utilized to ensure that the resulting adversarial samples maintain the essential abnormal behavior required for detection. The validation of our proposal was conducted using the WADI dataset obtained from a water distribution plant, providing a realistic industrial context. During validation, we assessed the mean error and the total number of adversarial samples generated by our approach, comparing it with the original Projected Gradient Descent method and the Carlini & Wagner attack across various parameter configurations. Remarkably, our proposal consistently achieved the best trade-off between mean error and the number of generated adversarial samples, showcasing its superiority in safeguarding industrial systems.