https://reunir.unir.net/handle/123456789/15167 Thu, 31 Oct 2024 11:56:05 GMT 2024-10-31T11:56:05Z https://reunir.unir.net/handle/123456789/15218 Editor’s Note Alonso, Ricardo S.; Chamoso, Pablo; Rodríguez-González, Sara; Novais, Paulo Research in Agents and Multiagent Systems has matured significantly in recent years, representing one of the main branches of Artificial Intelligence and currently there are numerous effective applications of these technologies combined with Deep Learning, Computer Vision or Natural Language Processing, including areas such as healthcare and Ambient Intelligence, smart cities and mobility, Industry 4.0, educational technology, and fintech, among many others. In this regard, the International Conference on Practical Applications of Agents and Multi-Agent System (PAAMS) provides an international forum to present and discuss the latest scientific advances and their effective applications in different sectors, evaluate the impact of the approach and facilitate technology transfer among different stakeholders. Currently, a series of co-located events specialized in different areas of research are held simultaneously with PAAMS, these being the International Congress on Blockchain and Applications (BLOCKCHAIN), the International Conference on Distributed Computing and Artificial Intelligence (DCAI), the International Conference on Decision Economics (DECON), the International Symposium on Ambient Intelligence (ISAmI), the International Conference on Methodologies and Intelligent Systems for Technology Enhanced Learning (MIS4TEL), and the International Conference on Practical Applications of Computational Biology & Bioinformatics (PACBB). In this regard, the present Special Issue includes a selection of extended papers presented at the 20th International Conference PAAMS 22 and its co-located events and held in L’Aquila (Italy), July 13-15, 2022. Submitted by Susana Figueroa Navarro (susana.figueroa.n@unir.net) on 2023-09-06T08:32:18Z No. of bitstreams: 1 ijimai8_3_0_0.pdf: 57556 bytes, checksum: f0b02c6bc1df1b55b4db509d8c856e50 (MD5); Made available in DSpace on 2023-09-06T08:32:18Z (GMT). No. of bitstreams: 1 ijimai8_3_0_0.pdf: 57556 bytes, checksum: f0b02c6bc1df1b55b4db509d8c856e50 (MD5) https://reunir.unir.net/handle/123456789/15218 https://reunir.unir.net/handle/123456789/15217 Violence Detection in Audio: Evaluating the Effectiveness of Deep Learning Models and Data Augmentation Durães, Dalila; Veloso, Bruno; Novais, Paulo Human nature is inherently intertwined with violence, impacting the lives of numerous individuals. Various forms of violence pervade our society, with physical violence being the most prevalent in our daily lives. The study of human actions has gained significant attention in recent years, with audio (captured by microphones) and video (captured by cameras) being the primary means to record instances of violence. While video requires substantial processing capacity and hardware-software performance, audio presents itself as a viable alternative, offering several advantages beyond these technical considerations. Therefore, it is crucial to represent audio data in a manner conducive to accurate classification. In the context of violence in a car, specific datasets dedicated to this domain are not readily available. As a result, we had to create a custom dataset tailored to this particular scenario. The purpose of curating this dataset was to assess whether it could enhance the detection of violence in car-related situations. Due to the imbalanced nature of the dataset, data augmentation techniques were implemented. Existing literature reveals that Deep Learning (DL) algorithms can effectively classify audio, with a commonly used approach involving the conversion of audio into a mel spectrogram image. Based on the results obtained for that dataset, the EfficientNetB1 neural network demonstrated the highest accuracy (95.06%) in detecting violence in audios, closely followed by EfficientNetB0 (94.19%). Conversely, MobileNetV2 proved to be less capable in classifying instances of violence. Submitted by Susana Figueroa Navarro (susana.figueroa.n@unir.net) on 2023-09-06T08:26:38Z No. of bitstreams: 1 ijimai8_3_7.pdf: 714559 bytes, checksum: a445d93fee207713f024d19b6bf6ab13 (MD5); Made available in DSpace on 2023-09-06T08:26:38Z (GMT). No. of bitstreams: 1 ijimai8_3_7.pdf: 714559 bytes, checksum: a445d93fee207713f024d19b6bf6ab13 (MD5) https://reunir.unir.net/handle/123456789/15217 https://reunir.unir.net/handle/123456789/15216 An Investigation Into Different Text Representations to Train an Artificial Immune Network for Clustering Texts Ferraria, Matheus A.; Ferraria, Vinicius A.; de Castro, Leandro N. Extracting knowledge from text data is a complex task that is usually performed by first structuring the texts and then applying machine learning algorithms, or by using specific deep architectures capable of dealing directly with the raw text data. The traditional approach to structure texts is called Bag of Words (BoW) and consists of transforming each word in a document into a dimension (variable) in the structured data. Another approach uses grammatical classes to categorize the words and, thus, limit the dimension of the structured data to the number of grammatical categories. Another form of structuring text data for analysis is by using a distributed representation of words, sentences, or documents with methods like Word2Vec, Doc2Vec, and SBERT. This paper investigates four classes of text structuring methods to prepare documents for being clustered by an artificial immune system called aiNet. The goal is to assess the influence of each structuring method in the quality of the clustering obtained by the system and how methods that belong to the same type of representation differ from each other, for example both LIWC and MRC are considered grammarbased models but each one of them uses completely different dictionaries to generate its representation. By using internal clustering measures, our results showed that vector space models, on average, presented the best results for the datasets chosen, followed closely by the state of the art SBERT model, and MRC had the overall worst performance. We could also observe a consistency in the number of clusters generated by each representation and for each dataset, having SBERT as the model that presented a number of clusters closer to the original number of classes in the data. Submitted by Susana Figueroa Navarro (susana.figueroa.n@unir.net) on 2023-09-06T08:20:25Z No. of bitstreams: 1 ijimai8_3_5.pdf: 353722 bytes, checksum: 84e1c9e34aff07d64e231c000192cdfd (MD5); Made available in DSpace on 2023-09-06T08:20:25Z (GMT). No. of bitstreams: 1 ijimai8_3_5.pdf: 353722 bytes, checksum: 84e1c9e34aff07d64e231c000192cdfd (MD5) https://reunir.unir.net/handle/123456789/15216 https://reunir.unir.net/handle/123456789/15215 Pollutant Time Series Analysis for Improving Air-Quality in Smart Cities López-Blanco, Raúl; Chaveinte García, Miguel; Alonso, Ricardo S.; Prieto, Javier; Corchado, Juan M. The evolution towards Smart Cities is the process that many urban centers are following in their quest for efficiency, resource optimization and sustainable growth. This step forward in the continuous improvement of cities is closely linked to the quality of life they want to offer their citizens. One of the key issues that can have the greatest impact on the quality of life of all city dwellers is the quality of the air they breathe, which can lead to illnesses caused by pollutants in the air. The application of new technologies, such as the Internet of Things, Big Data and Artificial Intelligence, makes it possible to obtain increasingly abundant and accurate data on what is happening in cities, providing more information to take informed action based on scientific data. This article studies the evolution of pollutants in the main cities of Castilla y León, using Generative Additive Models (GAM), which have proven to be the most efficient for making predictions with detailed historical data and which have very strong seasonalities. The results of this study conclude that during the COVID-19 pandemic containment period, there was an overall reduction in the concentration of pollutants. Submitted by Susana Figueroa Navarro (susana.figueroa.n@unir.net) on 2023-09-06T07:56:12Z No. of bitstreams: 1 ijimai8_3_9.pdf: 2271859 bytes, checksum: d134b99201a4bd287f347a2fe723b4ea (MD5); Made available in DSpace on 2023-09-06T07:56:12Z (GMT). No. of bitstreams: 1 ijimai8_3_9.pdf: 2271859 bytes, checksum: d134b99201a4bd287f347a2fe723b4ea (MD5) https://reunir.unir.net/handle/123456789/15215 https://reunir.unir.net/handle/123456789/15214 Consensus-Based Learning for MAS: Definition, Implementation and Integration in IVEs Carrascosa, C.; Enguix, F.; Rebollo, M.; Rincon, J. One of the main advancements in distributed learning may be the idea behind Google’s Federated Learning (FL) algorithm. It trains copies of artificial neural networks (ANN) in a distributed way and recombines the weights and biases obtained in a central server. Each unit maintains the privacy of the information since the training datasets are not shared. This idea perfectly fits a Multi-Agent System, where the units learning and sharing the model are agents. FL is a centralized approach, where a server is in charge of receiving, averaging and distributing back the models to the different units making the learning process. In this work, we propose a truly distributed learning process where all the agents have the same role in the system. We suggest using a consensus-based learning algorithm that we call Co-Learning. This process uses a consensus process to share the ANN models each agent learns using its private data and calculates the aggregated model. Co-Learning, as a consensus-based algorithm, calculates the average of the ANN models shared by the agents with their local neighbors. This iterative process converges to the averaged ANN model as a central server does. Apart from the definition of the Co-Learning algorithm, the paper presents its integration in SPADE agents, along with a framework called FIVE allowing to develop Intelligent Virtual Environments for SPADE agents. This framework has been used to test the execution of SPADE agents using Co-Learning algorithm in a simulation of an orange orchard field. Submitted by Susana Figueroa Navarro (susana.figueroa.n@unir.net) on 2023-09-06T07:34:30Z No. of bitstreams: 1 ijimai8_3_2.pdf: 958756 bytes, checksum: 8d28398827dd5615483a56b7b414301b (MD5); Made available in DSpace on 2023-09-06T07:34:30Z (GMT). No. of bitstreams: 1 ijimai8_3_2.pdf: 958756 bytes, checksum: 8d28398827dd5615483a56b7b414301b (MD5) https://reunir.unir.net/handle/123456789/15214 https://reunir.unir.net/handle/123456789/15213 Development of an Intelligent Classifier Model for Denial of Service Attack Detection Michelena, Álvaro; Aveleira-Mata, Jose; Jove, Esteban; Alaiz-Moretón, Héctor; Quintián, Héctor; Calvo-Rolle, José Luis The prevalence of Internet of Things (IoT) systems deployment is increasing across various domains, from residential to industrial settings. These systems are typically characterized by their modest computationa requirements and use of lightweight communication protocols, such as MQTT. However, the rising adoption of IoT technology has also led to the emergence of novel attacks, increasing the susceptibility of these systems to compromise. Among the different attacks that can affect the main IoT protocols are Denial of Service attacks (DoS). In this scenario, this paper evaluates the performance of six supervised classification techniques (Decision Trees, Multi-layer Perceptron, Random Forest, Support Vector Machine, Fisher Linear Discriminant and Bernoulli and Gaussian Naive Bayes) combined with the Principal Component Analysis (PCA) feature extraction method for detecting DoS attacks in MQTT networks. For this purpose, a real dataset containing all the traffic generated in the network and many attacks executed has been used. The results obtained with several models have achieved performances above 99% AUC. Submitted by Susana Figueroa Navarro (susana.figueroa.n@unir.net) on 2023-09-06T07:20:37Z No. of bitstreams: 1 ijimai8_3_3.pdf: 460755 bytes, checksum: 7a24d6bbdd35f105e746145de760bdc5 (MD5); Made available in DSpace on 2023-09-06T07:20:37Z (GMT). No. of bitstreams: 1 ijimai8_3_3.pdf: 460755 bytes, checksum: 7a24d6bbdd35f105e746145de760bdc5 (MD5) https://reunir.unir.net/handle/123456789/15213 https://reunir.unir.net/handle/123456789/15212 Automatic Cell Counting With YOLOv5: A Fluorescence Microscopy Approach López Flórez, Sebastián; González-Briones, Alfonso; Hernández, Guillermo; Ramos, Carlos; de la Prieta, Fernando Counting cells in a Neubauer chamber on microbiological culture plates is a laborious task that depends on technical experience. As a result, efforts have been made to advance computer vision-based approaches, increasing efficiency and reliability through quantitative analysis of microorganisms and calculation of their characteristics, biomass concentration, and biological activity. However, variability that still persists in these processes poses a challenge that is yet to be overcome. In this work, we propose a solution adopting a YOLOv5 network model for automatic cell recognition and counting in a case study for laboratory cell detection using images from a CytoSMART Exact FL microscope. In this context, a dataset of 21 expert-labeled cell images was created, along with an extra Sperm DetectionV dataset of 1024 images for transfer learning. The dataset was trained using the pretrained YOLOv5 algorithm with the Sperm DetectionV database. A laboratory test was also performed to confirm result’s viability. Compared to YOLOv4, the current YOLOv5 model had accuracy, precision, recall, and F1 scores of 92%, 84%, 91%, and 87%, respectively. The YOLOv5 algorithm was also used for cell counting and compared to the current segmentation-based U-Net and OpenCV model that has been implemented. In conclusion, the proposed model successfully recognizes and counts the different types of cells present in the laboratory. Submitted by Susana Figueroa Navarro (susana.figueroa.n@unir.net) on 2023-09-06T07:08:19Z No. of bitstreams: 1 ijimai8_3_6.pdf: 697392 bytes, checksum: c1f8f47261458de62606dcb4d737418c (MD5); Made available in DSpace on 2023-09-06T07:08:19Z (GMT). No. of bitstreams: 1 ijimai8_3_6.pdf: 697392 bytes, checksum: c1f8f47261458de62606dcb4d737418c (MD5) https://reunir.unir.net/handle/123456789/15212 https://reunir.unir.net/handle/123456789/15211 A Survey on Demand-Responsive Transportation for Rural and Interurban Mobility Martí, Pasqual; Jordán, Jaume; González Arrieta, María Angélica; Julian, Vicente Rural areas have been marginalized when it comes to flexible, quality transportation research. This review article brings together papers that discuss, analyze, model, or experiment with demand-responsive transportation systems applied to rural settlements and interurban transportation, discussing their general feasibility as well as the most successful configurations. For that, demand-responsive transportation is characterized and the techniques used for modeling and optimization are described. Then, a classification of the relevant publications is presented, splitting the contributions into analytical and experimental works. The results of the classification lead to a discussion that states open issues within the topic: replacement of public transportation with demandresponsive solutions, disconnection between theoretical and experimental works, user-centered design and its impact on adoption rate, and a lack of innovation regarding artificial intelligence implementation on the proposed systems. Submitted by Susana Figueroa Navarro (susana.figueroa.n@unir.net) on 2023-09-05T16:15:21Z No. of bitstreams: 1 ijimai8_3_4.pdf: 412656 bytes, checksum: f18ea08635e89eb0ce496ee4f982c51f (MD5); Made available in DSpace on 2023-09-05T16:15:21Z (GMT). No. of bitstreams: 1 ijimai8_3_4.pdf: 412656 bytes, checksum: f18ea08635e89eb0ce496ee4f982c51f (MD5) https://reunir.unir.net/handle/123456789/15211 https://reunir.unir.net/handle/123456789/15198 Using Large Language Models to Shape Social Robots’ Speech Sevilla-Salcedo, Javier; Fernádez-Rodicio, Enrique; Martín-Galván, Laura; Castro-González, Álvaro; Castillo, José C.; Salichs, Miguel A. Social robots are making their way into our lives in different scenarios in which humans and robots need to communicate. In these scenarios, verbal communication is an essential element of human-robot interaction. However, in most cases, social robots’ utterances are based on predefined texts, which can cause users to perceive the robots as repetitive and boring. Achieving natural and friendly communication is important for avoiding this scenario. To this end, we propose to apply state-of- the-art natural language generation models to provide our social robots with more diverse speech. In particular, we have implemented and evaluated two mechanisms: a paraphrasing module that transforms the robot’s utterances while keeping their original meaning, and a module to generate speech about a certain topic that adapts the content of this speech to the robot’s conversation partner. The results show that these models have great potential when applied to our social robots, but several limitations must be considered. These include the computational cost of the solutions presented, the latency that some of these models can introduce in the interaction, the use of proprietary models, or the lack of a subjective evaluation that complements the results of the tests conducted. Submitted by Susana Figueroa Navarro (susana.figueroa.n@unir.net) on 2023-09-04T16:02:13Z No. of bitstreams: 1 ijimai8_3_1.pdf: 440548 bytes, checksum: 5eb947a869e7c9bb1f258fde803c3198 (MD5); Made available in DSpace on 2023-09-04T16:02:13Z (GMT). No. of bitstreams: 1 ijimai8_3_1.pdf: 440548 bytes, checksum: 5eb947a869e7c9bb1f258fde803c3198 (MD5) https://reunir.unir.net/handle/123456789/15198 https://reunir.unir.net/handle/123456789/15197 Problem Detection in the Edge of IoT Applications Bernabé-Sánchez, Iván; Fernández, Alberto; Billhardt, Holger; Ossowski, Sascha Due to technological advances, Internet of Things (IoT) systems are becoming increasingly complex. They are characterized by being multi-device and geographically distributed, which increases the possibility of errors of different types. In such systems, errors can occur anywhere at any time and fault tolerance becomes an essential characteristic to make them robust and reliable. This paper presents a framework to manage and detect errors and malfunctions of the devices that compose an IoT system. The proposed solution approach takes into account both, simple devices such as sensors or actuators, as well as computationally intensive devices which are distributed geographically. It uses knowledge graphs to model the devices, the system’s topology, the software deployed on each device and the relationships between the different elements. The proposed framework retrieves information from log messages and processes this information automatically to detect anomalous situations or malfunctions that may affect the IoT system. This work also presents the ECO ontology to organize the IoT system information. Submitted by Susana Figueroa Navarro (susana.figueroa.n@unir.net) on 2023-09-04T15:53:52Z No. of bitstreams: 1 ijimai8_3_8_0.pdf: 896555 bytes, checksum: 147dbd5992157eb21e49bc88dde04a02 (MD5); Made available in DSpace on 2023-09-04T15:53:52Z (GMT). No. of bitstreams: 1 ijimai8_3_8_0.pdf: 896555 bytes, checksum: 147dbd5992157eb21e49bc88dde04a02 (MD5) https://reunir.unir.net/handle/123456789/15197