Emilio S. Corchado’s research while affiliated with University of Salamanca and other places

Automated disease diagnosis in plants is crucial for the agriculture industry to maintain crop health and increase yields, which has significant implications for global food security and the economy. The use of convolutional neural networks (CNNs) for disease diagnosis has gained much attention due to their ability to detect patterns in the images of plant leaves, allowing for the accurate diagnosis of diseases. However, one of the major challenges in using CNNs is their limited interpretability, which makes it difficult to understand the reasoning behind the model’s output. In this work, we explore the potential of CNNs for plant disease diagnosis and propose the use of explainable artificial intelligence (XAI) methods to improve the interpretability of the CNN’s output. We first discuss the state of the art in plant disease detection, including conventional methods such as Polymerase Chain Reaction (PCR) and Isothermal Amplification based diagnosis, and the rise of CNNs in the field. We then introduce the architecture and the principles of CNNs, highlighting their ability to classify images and their use in plant disease diagnosis. However, due to their black-box nature, the interpretation of CNN outputs remains challenging. To address this, we propose the use of post-hoc XAI methods, specifically LIME, SHAP, and Grad-CAM, to provide insights into CNN’s decision-making process. Our study aims to demonstrate the potential of CNNs for plant disease diagnosis and the importance of interpretability in deep learning models. We hope our work will contribute to the development of more accurate and interpretable CNN models for disease diagnosis in plants, ultimately leading to more efficient and sustainable agriculture practices.KeywordsConvolutional Neural Network (CNN)Deep LearningInterpretabilityeXplainable Artificial Intelligence (XAI)

The increasing use of artificial intelligence (AI) in various domains has led to a growing need for AI systems to provide interpretable and understandable results. This need has given rise to the field of explainable artificial intelligence (XAI). XAI refers to the development of AI systems that can provide a clear and interpretable explanation of their decision-making processes to the end-users. In this paper, we provide a comprehensive review of the state-of-the-art techniques in XAI. We start by proposing a classification of the different XAI techniques, from the moment of application to the extent of the explanation and other specific properties of the methods. We then review multiple XAI methods, including ad-hoc techniques, local and global explanations, and other subgroups. Not only the theory behind them is explained, but also their practical application, so as to show the different outputs that can be obtained with different python implementations. Finally, we conclude the paper by highlighting the future lines of research in XAI and its potential impact on society.KeywordsExplainable Artificial Intelligence (XAI)ProtoDashDIP-VAESHAPLIMETREPANProfWeightTED

Energy aggregation business models play a crucial role in maximizing the benefits of renewable energy generation and consumption. In this context, this paper presents an examination of energy system models and their correlation with aggregation business models, emphasizing algorithmic approaches. The distinct characteristics of energy system models in comparison to traditional business models are elucidated. The study encompasses diverse model types employed in the energy sector, including optimization, classification, clustering, and integrated assessment models. Additionally, it investigates the algorithms utilized in aggregation business models, such as demand response, virtual power plants, and peer-to-peer energy trading, highlighting both their advantages and challenges. The paper concludes that the advancement of energy aggregation business models holds significant promise in addressing the intricate complexities of contemporary power systems and the ongoing energy transition. Future research endeavors involve the exploration of Electroencephalography (EEG) techniques within energy aggregation models, inspired by the information processing mechanisms of the human brain.KeywordsAggregationEnergy CommunitiesSmart GridBusiness Model

Industry 4.0 (the Fourth Industrial Revolution) is a concept devised for improving the operation of modern factories through the use of the latest technologies, under paradigms such as the Industrial Internet of Things (IIoT) or Big Data. One of such technologies is Blockchain, which is able to provide industrial processes with security, trust, traceability, reliability and automation. This paper proposes a technological framework that combines an information sharing platform and a Blockchain platform. One of the main features of the this framework is the use of smart contracts for validating and auditing the content received throughout the production process to ensure the correct traceability of the data. The conclusion drawn from this study is that this technology is under-researched and has significant potential to support and enhance the industrial revolution. Moreover, this study identifies areas for future research.

One of the biggest challenges in training supervised models is the lack of amount of labeled data for training the model and facing overfitting and underfitting problems. One of the solutions for solving this problem is data augmentation. There have been many developments in data augmentation of the image files, especially in medical image type datasets, by doing some changes on the original file such as Random cropping, Filliping, Rotating, and so on, in order to make a new sample file. Or use Deep Learning models to generate similar samples like Generative Adversarial Networks, Convolutional Neural Networks and so on. However, in numerical dataset, there have not been enough advances. In this paper, we are proposing to use the Gaussian Mixture Models (GMMs) to augment more data very similar to the original Numerical dataset. The results demonstrated that the Mean Absolute Error decreases meaning that the regression model became more accurate.

The wide acceptance by people when using mobile applications to communicate or interact, replacing traditional media, makes mobile applications a very useful tool today. Thanks to the interconnection that smartphones provide and the technological evolution that allows everyday objects to be connected with people, the concept of smart cities has gained strength. This paper presents an application that aims to connect the citizens of a city in a simple way with the administration and authorities of different sectors, in order to speed up communication and provide cities with a fast route (practically in real time) with which to obtain feedback from citizens.