Special Issue "Advances in Explainable Artificial Intelligence and Edge Computing Applications"

A special issue of Electronics (ISSN 2079-9292). This special issue belongs to the section "Computer Science & Engineering".

Deadline for manuscript submissions: closed (31 January 2022) | Viewed by 4578

Special Issue Editors

Prof. Dr. Juan M. Corchado
E-Mail Website
Guest Editor
BISITE Research Group, Edificio Multiusos I+D+i, University of Salamanca, 37007 Salamanca, Spain
Interests: artificial Intelligence; machine learning; edge computing; distributed computing; Blockchain; consensus model; smart cities; smart grid
Special Issues, Collections and Topics in MDPI journals
Prof. Dr. Sascha Ossowski
E-Mail Website
Guest Editor
Universidad Rey Juan Carlos, Centre for Intelligent Information Technologies (CETINIA), Mostoles, Spain
Interests: software systems; artificial intelligence

Special Issue Information

Dear Colleagues,

Artificial Intelligence (AI) and its applications have undergone remarkable experimental development in the last decade and are now the basis for a large number of decision support systems. There are countless models and algorithms, with different typologies, and their applications are highly varied in terms of both configuration and use cases. In addition, there has been an increase in the number of tools that simplify the application of all the existing algorithms, and there are powerful data visualization systems. Although the success of AI has been demonstrated in many models, in many cases, these models lack mechanisms to explain why a given result is obtained. Many algorithms continue to be black boxes—they receive data and deliver results, but they do not provide any explanation.  

This Special Issue focuses on the concept of Explainable Artificial Intelligence (XAI). These models are applied to systems that complement each other, facilitating the parallel use of data treatment and knowledge processing algorithms. This makes it possible to simultaneously process both relational and non-relational data from databases and sources that generate data in real time, such as IoT sensors. To this end, the use of data analysis systems with AI algorithms and the parallel use of mathematical models for the creation of self-explanatory expert mixing models that incorporate, for example, deep symbolic learning, convolutional neural networks, compartmental mathematical models, dynamic data assimilation models, fuzzy systems, Bayesian networks, etc. as well as other models from the field of machine learning are of interest. 

This Special Issue invites papers on all those research approaches focused on the relationship between evolution in the area of transport with a high incidence in the environment from the perspective of efficiency, which has become one of the neuralgic centers of sustainability. The aim of the issue is to encourage the better production, consumption, and movement of goods, as well as the movement of people with fewer resources and a smaller environmental impact. 

The topics of interest in this issue include but are not limited to:

  • Distributed problem solving;
  • Agent cooperation and negotiation;
  • Human agent interaction, social networks, virtual communities;
  • Reputation, trust, privacy, and security;
  • Agent engineering and development tools;
  • Internet of Things, sensors, and actuators;
  • Big data and machine learning;
  • Smart cities, smart homes, smart buildings, smart health, smart mobility, and transportation;
  • Semantic web, linked data;
  • Expert systems;
  • Edge computing, Machine learning, etc.;
  • Deep intelligence. 

This Special Issue concerns extended research presented during the set of the following conferences: PAAMS, DCAI, ISAmI, PACBB, MIS4TEL, and BLOCKCHAIN.

Prof. Dr. Juan M. Corchado
Prof. Dr. Sascha Ossowski
Prof. Dr. Sara Rodriguez
Prof. Dr. Fernando De la Prieta
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Electronics is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2000 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

Keywords

  • Explainable Artificial Intelligence
  • Deep Learning
  • Agent engineering

Published Papers (9 papers)

Order results
Result details
Select all
Export citation of selected articles as:

Research

Article
Toward Autonomous and Distributed Intersection Management with Emergency Vehicles
Electronics 2022, 11(7), 1089; https://doi.org/10.3390/electronics11071089 - 30 Mar 2022
Viewed by 262
Abstract
Numerous approaches have attempted to develop systems that more appropriately manage street crossings in cities in recent years. Solutions range from intelligent traffic lights to complex, centralized protocols that evaluate the policies that vehicles must comply with at intersections. Such works attempt to [...] Read more.
Numerous approaches have attempted to develop systems that more appropriately manage street crossings in cities in recent years. Solutions range from intelligent traffic lights to complex, centralized protocols that evaluate the policies that vehicles must comply with at intersections. Such works attempt to provide traffic-control strategies at intersections where the complexity of a dynamic environment, with vehicles crossing in different directions and multiple conflict points, pose a significant challenge for city traffic optimization. Traditionally, a traffic-control system at an intersection gives the green light to one lane while keeping the other lanes on red. But there may be situations in which there are different levels of vehicle priority; for example, emergency vehicles may have priority at intersections. Thus, this work proposes a distributed junction-management protocol that pays special attention to emergency vehicles. The proposed algorithm implements rules based on the distributed intersection management (DIM) protocol; such rules are used by vehicles while negotiating their crossing through the intersection. The proposal also seeks to affect the traffic flow of non-priority vehicles minimally. An evaluation and comparison of the proposed algorithm are presented in the paper. Full article
Show Figures

Figure 1

Article
Weakness Evaluation on In-Vehicle Violence Detection: An Assessment of X3D, C2D and I3D against FGSM and PGD
Electronics 2022, 11(6), 852; https://doi.org/10.3390/electronics11060852 - 09 Mar 2022
Viewed by 394
Abstract
When constructing a deep learning model for recognizing violence inside a vehicle, it is crucial to consider several aspects. One aspect is the computational limitations, and the other is the deep learning model architecture chosen. Nevertheless, to choose the best deep learning model, [...] Read more.
When constructing a deep learning model for recognizing violence inside a vehicle, it is crucial to consider several aspects. One aspect is the computational limitations, and the other is the deep learning model architecture chosen. Nevertheless, to choose the best deep learning model, it is necessary to test and evaluate the model against adversarial attacks. This paper presented three different architecture models for violence recognition inside a vehicle. These model architectures were evaluated based on adversarial attacks and interpretability methods. An analysis of the model’s convergence was conducted, followed by adversarial robustness for each model and a sanity-check based on interpretability analysis. It compared a standard evaluation for training and testing data samples with the adversarial attacks techniques. These two levels of analysis are essential to verify model weakness and sensibility regarding the complete video and in a frame-by-frame way. Full article
Show Figures

Figure 1

Article
Deep Learning for Activity Recognition Using Audio and Video
Electronics 2022, 11(5), 782; https://doi.org/10.3390/electronics11050782 - 03 Mar 2022
Viewed by 304
Abstract
Neural networks have established themselves as powerhouses in what concerns several types of detection, ranging from human activities to their emotions. Several types of analysis exist, and the most popular and successful is video. However, there are other kinds of analysis, which, despite [...] Read more.
Neural networks have established themselves as powerhouses in what concerns several types of detection, ranging from human activities to their emotions. Several types of analysis exist, and the most popular and successful is video. However, there are other kinds of analysis, which, despite not being used as often, are still promising. In this article, a comparison between audio and video analysis is drawn in an attempt to classify violence detection in real-time streams. This study, which followed the CRISP-DM methodology, made use of several models available through PyTorch in order to test a diverse set of models and achieve robust results. The results obtained proved why video analysis has such prevalence, with the video classification handily outperforming its audio classification counterpart. Whilst the audio models attained on average 76% accuracy, video models secured average scores of 89%, showing a significant difference in performance. This study concluded that the applied methods are quite promising in detecting violence, using both audio and video. Full article
Show Figures

Figure 1

Article
Anomaly Detection on Natural Language Processing to Improve Predictions on Tourist Preferences
Electronics 2022, 11(5), 779; https://doi.org/10.3390/electronics11050779 - 03 Mar 2022
Viewed by 398
Abstract
Argumentation-based dialogue models have shown to be appropriate for decision contexts in which it is intended to overcome the lack of interaction between decision-makers, either because they are dispersed, they are too many, or they are simply not even known. However, to support [...] Read more.
Argumentation-based dialogue models have shown to be appropriate for decision contexts in which it is intended to overcome the lack of interaction between decision-makers, either because they are dispersed, they are too many, or they are simply not even known. However, to support decision processes with argumentation-based dialogue models, it is necessary to have knowledge of certain aspects that are specific to each decision-maker, such as preferences, interests, and limitations, among others. Failure to obtain this knowledge could ruin the model’s success. In this work, we sought to facilitate the information acquisition process by studying strategies to automatically predict the tourists’ preferences (ratings) in relation to points of interest based on their reviews. We explored different Machine Learning methods to predict users’ ratings. We used Natural Language Processing strategies to predict whether a review is positive or negative and the rating assigned by users on a scale of 1 to 5. We then applied supervised methods such as Logistic Regression, Random Forest, Decision Trees, K-Nearest Neighbors, and Recurrent Neural Networks to determine whether a tourist likes/dislikes a given point of interest. We also used a distinctive approach in this field through unsupervised techniques for anomaly detection problems. The goal was to improve the supervised model in identifying only those tourists who truly like or dislike a particular point of interest, in which the main objective is not to identify everyone, but fundamentally not to fail those who are identified in those conditions. The experiments carried out showed that the developed models could predict with high accuracy whether a review is positive or negative but have some difficulty in accurately predicting the rating assigned by users. Unsupervised method Local Outlier Factor improved the results, reducing Logistic Regression false positives with an associated cost of increasing false negatives. Full article
Show Figures

Figure 1

Article
Developing IoT Artifacts in a MAS Platform
Electronics 2022, 11(4), 655; https://doi.org/10.3390/electronics11040655 - 19 Feb 2022
Cited by 1 | Viewed by 364
Abstract
The Internet of Things (IoT) is a growing computational paradigm where all kinds of everyday objects are interconnected, forming a vast cyberphysical environment at the edge between the virtual and the real world. Since the emergence of the IoT, Multi-Agent Systems (MAS) technology [...] Read more.
The Internet of Things (IoT) is a growing computational paradigm where all kinds of everyday objects are interconnected, forming a vast cyberphysical environment at the edge between the virtual and the real world. Since the emergence of the IoT, Multi-Agent Systems (MAS) technology has been successfully applied in this area, proving itself to be an appropriate paradigm for developing distributed, intelligent systems containing sets of IoT devices. However, this technology still lacks effective mechanisms to integrate the enormous diversity of existing IoT devices systematically. In this context, this paper introduces the concept of the IoT artifact as a new interface abstraction for the development of MAS based on IoT devices. The IoT artifact strictly conforms to the Agents and Artifacts (A&A) meta-model, and it also adopts the programming model of the SPADE multi-agent platform, providing both a consistent theoretical framework and a practical model for real-world applications. Full article
Show Figures

Figure 1

Article
Processing at the Edge: A Case Study with an Ultrasound Sensor-Based Embedded Smart Device
Electronics 2022, 11(4), 550; https://doi.org/10.3390/electronics11040550 - 11 Feb 2022
Viewed by 299
Abstract
In the current context of the Internet of Things, embedded devices can have some intelligence and distribute both data and processed information. This article presents the paradigm shift from a hierarchical pyramid to an inverted pyramid that is the basis for edge, fog, [...] Read more.
In the current context of the Internet of Things, embedded devices can have some intelligence and distribute both data and processed information. This article presents the paradigm shift from a hierarchical pyramid to an inverted pyramid that is the basis for edge, fog, and cloud-based architectures. To support the new paradigm, the article presents a distributed modular architecture. The devices are made up of essential elements, called control nodes, which can communicate to enhance their functionality without sending raw data to the cloud. To validate the architecture, identical control nodes equipped with a distance sensor have been implemented. Each module can read the distance to each vehicle and process these data to provide the vehicle’s speed and length. In addition, the article describes how connecting two or more CNs, forming an intelligent device, can increase the accuracy of the parameters measured. Results show that it is possible to reduce the processing load up to 22% in the case of sharing processed information instead of raw data. In addition, when the control nodes collaborate at the edge level, the relative error obtained when measuring the speed and length of a vehicle is reduced by one percentage point. Full article
Show Figures

Graphical abstract

Article
Unpacking the Role of Artificial Intelligence for a Multimodal Service System Design
Electronics 2022, 11(4), 549; https://doi.org/10.3390/electronics11040549 - 11 Feb 2022
Viewed by 297
Abstract
Since requirements of service demands are becoming increasingly complex and diversified, one of the success factors of a multimodal service system is its capability to design a specific service instance satisfying a specific set of requirements. This capability is further highlighted in Ad [...] Read more.
Since requirements of service demands are becoming increasingly complex and diversified, one of the success factors of a multimodal service system is its capability to design a specific service instance satisfying a specific set of requirements. This capability is further highlighted in Ad Hoc Multimodal Service Systems (AHMSSs), where service instances rarely follow a standard form of service delivery and exist only for a limited time. However, due to the increasing scale and frequency of services in many business and public sectors, meeting the desired level of capability has become troublesome. A well-designed Artificial Intelligence (AI) approach can be a solution to the difficulty by addressing the underlying complexity and uncertainty of the AHMSS design process. To conceptualize and foster AI applications to an AHMSS, this study identifies key decision-making problems in the AHMSS design process and discusses the role of AI in the process. The results will form the basis for AI development and implementation for an AHMSS and relevant service systems. Full article
Show Figures

Figure 1

Article
Comparison of Predictive Models with Balanced Classes Using the SMOTE Method for the Forecast of Student Dropout in Higher Education
Electronics 2022, 11(3), 457; https://doi.org/10.3390/electronics11030457 - 03 Feb 2022
Viewed by 407
Abstract
Based on the premise that university student dropout is a social problem in the university ecosystem of any country, technological leverage is a way that allows us to build technological proposals to solve a poorly met need in university education systems. Under this [...] Read more.
Based on the premise that university student dropout is a social problem in the university ecosystem of any country, technological leverage is a way that allows us to build technological proposals to solve a poorly met need in university education systems. Under this scenario, the study presents and analyzes eight predictive models to forecast university dropout, based on data mining methods and techniques, using WEKA for its implementation, with a dataset of 4365 academic records of students from the National University of Moquegua (UNAM), Peru. The objective is to determine which model presents the best performance indicators to forecast and prevent student dropout. The study aims to propose and compare the accuracy of eight predictive models with balanced classes, using the SMOTE method for the generation of synthetic data. The results allow us to confirm that the predictive model based on Random Forest is the one that presents the highest accuracy and robustness. This study is of great interest to the educational community as it allows for predicting the possible dropout of a student from a university career and being able to take corrective actions both at a global and individual level. The results obtained are highly interesting for the university in which the study has been carried out, obtaining results that generally outperform the results obtained in related works. Full article
Show Figures

Figure 1

Article
Real Business Applications and Investments in Blockchain Technology
Electronics 2022, 11(3), 438; https://doi.org/10.3390/electronics11030438 - 01 Feb 2022
Viewed by 767
Abstract
This paper provides an empirical study to identify the objective of companies that are currently investing in adopting blockchain technologies to improve their processes and services. Unlike other studies based on the theoretical potential application of blockchain technology in different sectors, the main [...] Read more.
This paper provides an empirical study to identify the objective of companies that are currently investing in adopting blockchain technologies to improve their processes and services. Unlike other studies based on the theoretical potential application of blockchain technology in different sectors, the main objective of this paper is to analyze real projects and investment of companies in blockchain technology. More than 100 blockchain projects from different sectors were examined with the aim of extracting the perceived applicability and business value of blockchain technology by managers, customers, and partners. We identified the most demanded business value and functional properties in each sector and company size, as well as the relationship between the properties that are demanded together. This article assesses the main functional values attributed to blockchain, highlighting those really appreciated by companies that invest in them and identifying new applications of blockchain technology in different sectors, and generating organizational change. The article reveals that, as expected, significant deviations are already occurring between theoretical applications identified in the literature and those finally adopted by the industry. Full article
Show Figures

Figure 1

Back to TopTop