Associative classifiers have been widely used in many domains due to their inherent interpretability. They are built in steps, one of them aimed at ranking the rules, usually performed through objective measures. Works aim to modify this step in order to obtain a classifier with better performance. Among them are those that use multiple measures simultaneously in order to consider different points of view for a given rule. However, these works present problems regarding execution time and interpretability. Here we show the use of ranking aggregation methods, specifically Borda’s methods, to rank the rules through a set of measures. Our results demonstrate that our solution is fast to execute and still guarantee the interpretability of the models, since they contain a statistically significant smaller number of rules.
Due to the amount of software that is produced every day in the automotive industry, improve software quality became a necessity, especially in areas where safety is a critical point, such as autonomous driving. Following on from this, continuous inspection of software is important, timely and central, to guarantee software quality and avoid the worst scenarios related to the automotive industry, particularly with autonomous driving. Even so, this proposal is motivated by the huge number of failures associated with low quality software. In this way, the main objective of this research is to achieve a solution for continuous inspection of software quality in the context of the company Bosch Car Multimedia.
Facebook Papers refers to an investigative project, conducted by a consortium of seventeen US media outlets, that had access to thousands of internal Facebook documents the set of documents was leaked by a former employee and whistleblower. This article aims to understand the research that was conducted internally by Facebook teams focusing on Body image and appearance on Instagram. The methodology employed combines bibliographic and documentary research, with descriptive analysis of research reports on Body Image, appearance, and social comparison promoted by the research teams of the company Facebook and released internally between the years 2019, 2020 and 2021. The documents were made available in 2021 by The Wall Street Journal. Thus, this paper describes i) the methodologies employed and the research findings; ii) what suggestions were made for improvement in the safety and wellbeing of individuals on Instagram.
Consumers are increasingly aware of issues related to food safety and fraud, as well as environmental and ecological impacts due to food production. Besides the quality of the products, there is currently also a growing concern about the origin of the products and the path they take to reach the final consumers. Consumers are demanding transparency across the entire value chain of the products they consume. That’s why it’s necessary to design and develop traceability platforms for these products. The platform must be able to collect information at all stages of the supply chain and store it in a system external and shared by all operators in the value chain. This information must be available to all participants in the supply chain, including farmers, health authorities, government industries, sellers, retailer, final consumers, and any other actor involved in the value chain. This article describes the planning and implementation of a platform that allows to trace back and forth fruits and vegetables. This platform would be decentralized by using a blockchain-based approach. The proposed platform would provide an easy way to visualize information about products to any participant in this supply chain.
Virtual Learning Environments began to be adopted more frequently due to the demands generated by the pandemic and the need for new teaching and assessment methodologies, especially in Higher Education. Many evaluative practices today are directly associated with Online Teaching, and it is necessary to point out elements that indicate the appropriate choice of the Assessment Instrument in Online Teaching. The research aims to propose a distributed system that helps in the adequate indication of the choice of the Assessment Instrument that will be applied, based on the student’s Learning Style. To do so, we will use a Multiagent System that reasons and proposes the indication of the Assessment Instrument being closer to the student's Learning Style so that the best use of the results obtained in the assessment is obtained, contributing with metric indicators that represent the understanding of the skills and skills achieved. Thus, our study, still under development, is a Systematic Literature Review and seeks to survey the state of the art of research related to Online Assessment.
Metaverse is a compound word combining meta (beyond; transcending) and verse (the root of the universe; cosmos; the whole world), which denotes a new virtual universe created beyond the real world. This paper aims to explore the conceptual structure of the metaverse from a Web of Science (WoS) database search and metadata analysis using the R-Bibliometrix package. This research identified the following performance indicators: most relevant sources, corresponding authors’ country, most cited countries, word cloud extracted from WoS subject categories, and cumulative dynamics of authors’ keywords. The conceptual structure analysis used the following features: thematic map of authors’ keywords; thematic evolution of authors’ keywords; and factor analysis using the multiple correspondence analysis method. It is concluded that the metaverse will merge the virtual and physical worlds and create a new engagement model based on several emerging technologies, such as virtual and augmented reality, artificial intelligence, blockchain, and edge computing.
This article discusses the importance of hierarchy in the text for a full understanding. Natural language processing algorithms may be used to identify primary and secondary ideas, semantic objects, and their features and functions in texts. By following these steps from general concepts to more specific ones depending on the complexity and size of the text as well as its purpose, such algorithms allow for deeper analysis of textual content. However, full comprehension still requires human skills such as making inferences and interpreting figurative language.
This work presents the SkillsMe platform, a computational artifact that motivates its users to search for new knowledge. It is an application for mobile devices based on the Android system, having been developed using the Kotlin language. Its objective is to provide a collaborative environment for the exchange of knowledge among its users, allowing the sharing of knowledge, experiences and learning. Its target audience is made up of people interested in improving their intellectual level, through personal or professional development or a hobby. Two experimental evaluations were carried out to validate the SkillsMe platform, the first in the form of a questionnaire with essay questions, and the second in the form of an evaluation of the usability of the SkillsMe platform, using the System Usability Scale (SUS) method. The SkillsMe platform was good rated for its usability.
Games with purposes beyond entertainment, the so-called serious games, have been useful tools in professional training, especially in engaging participants. However, their evaluation and, also, their adaptable characteristics to different scenarios, audiences and contexts remain challenges. This paper examines the application of serious games in professional training, their results and adaptable ways to achieve certain goals. Using the Design Science Research (DSR) methodology, a framework was built to develop and evaluate serious games to improve user experience, learning outcomes, knowledge transfer to work situations, and the application of the skills practised in the game in real professional settings. At this stage, the investigation presents a framework regarding the triangulation of data collected from a systematic literature review, focus groups and interviews. Following the DSR methodology, the next steps of this investigation, listed at the end of the paper, are the demonstration of the framework in serious game development and the evaluation and validation of this artefact.
Metaverse is a compound word combining meta (beyond; transcending) and verse (the root of the universe; cosmos; the whole world), which denotes a new virtual universe created beyond the real world. This paper aims to explore the conceptual structure of the metaverse from a Web of Science (WoS) database search and metadata analysis using the R-Bibliometrix package. This research identified the following performance indicators: most relevant sources, corresponding authors’ country, most cited countries, word cloud extracted from WoS subject categories, and cumulative dynamics of authors’ keywords. The conceptual structure analysis used the following features: thematic map of authors’ keywords; thematic evolution of authors’ keywords; and factor analysis using the multiple correspondence analysis method. It is concluded that the metaverse will merge the virtual and physical worlds and create a new engagement model based on several emerging technologies, such as virtual and augmented reality, artificial intelligence, blockchain, and edge computing.
In the last decades, some research on online education has focused on better understanding a particular type of behavior gamed by students, known as gaming the system. In this article, we follow this track and present the first results of a new research approach to analyze and compare existing features commonly assigned to students used by automatic detectors for the behavior gaming the system in different situations, including problem-solving, considering various criteria. We have addressed the following main research question: What information or feature patterns about the students have been considered by automatic detector models of gaming the system behaviors? To answer this question, we have followed a knowledge discovery methodology in two studies: literature review and pattern identification. These studies were carried out in five concurrent stages: Selection of papers, Identification of the relevant works, Extraction of variables, Creating the patterns, and Mapping of the works in the standards. Results indicate that the detectors reported in the literature use eight variable patterns.
Deep learning techniques have recently gained increasing attention not only among computer science researchers but are also being applied in a wide range of fields. However, deep learning models demand huge amounts of data. Furthermore, fully supervised learning requires labeled data to solve classification, recognition, and segmentation problems. Data labeling and annotation in the medical domain are time-consuming and labor-intensive. Semi-supervised learning has demonstrated the ability to improve deep learning performance when labeled data is scarce. However, it is still an open and challenging question on how to leverage not only labeled data but also the huge amount of unlabeled data. In this paper, the problem of pancreatic cancer detection on CT scans is addressed by a semi-supervised learning approach based on pseudo-labeling. Preliminary results are promising and show the potential of semi-supervised deep learning to detect pancreatic cancer at an early stage with a limited amount of labeled data.
The objective of this research is to implement an application that allows the Clinic of the Universidad Nacional Mayor de San Marcos (UNMSM) to optimally manage the information and services offered to teachers, employees and students. This improvement involves the combination of technologies with tools and processes that will allow transforming the stored data from information to knowledge aligned with the strategies of the Admissions and General Management areas and is part of the business intelligence proposed by the UNMSM Clinic. The methodology used in the implementation was Hephaestus, which was selected after a "Benchmarking", the requirements were obtained in a survey that allowed to identify the Key Performance Indicators (KPI) to provide relevant information and improve the decision making of the UNMSM Clinic. As a result, the implemented application allows to generate reports that can be graphs and pivot tables related to each specified requirement, which are generated from the deployment of the OLAP cube whose base is composed of a multidimensional model of a Data Mart. Achieving efficiency in decision making.
In this article, the statistical behavior of the activation intensities of the action units that represent the micro-expressions of the facial expressions for four main emotions, happiness, anger, sadness, and surprise, is analyzed. Based on the results obtained, the distribution of each unit of action is modeled through probability density functions, which will allow the creation of an infinity of random samples, which contribute to emotion evaluation processes and especially to artificial intelligence techniques that require a high number of samples for their training processes.
This article presents a case study on the use of photogrammetry for capturing the reality of a hydroelectric power plant, specifically the Mascarenhas de Moraes plant, with the aim of generating a parametric model of its physical structures. The photogrammetric survey was conducted using a drone to capture images of the dam, spillways, and water intake, which were then processed using Agisoft Metashape and Autodesk software. The resulting point cloud was used to create a parametric model of the plant’s concrete structures in Revit. The study demonstrates the efficacy of photogrammetry in capturing the reality of complex structures such as hydroelectric power plants, and highlights the potential for further research into the use of this technique in the field of engineering.
In an age where citizens are constantly moving between different places, transport demand is extremely high, and so, it is important to have sophisticated public transportation systems in place to ensure a sustainable development of urban areas and meet the needs of citizens. Public transport operators consequently need to provide reliable services in order to minimize disruption events that can affect the vehicles and their drivers, such as breakdowns, accidents or illnesses. The project here described focuses on the type of events and approaches related with the vehicle drivers and the identification of both their performance profiles and health condition while in operation. For that purpose, existing nonintrusive technologies present on the vehicle are leveraged, able to collect data related to physiological measurements taken in realtime. Such sensitive data will be processed, stored and shared in a secure manner, using blockchain-based technologies, so that only authenticated and authorized parties will be able to access the data, according to their clearance level, through an Application Programming Interface (API) designed for that purpose. The architecture of the system will be microservices-based, with components deployed at different infrastructure levels—from On Board Units (OBUs) in vehicles up to cloud-based subsystems.
Context: Glass Containers’ International Manufacturing. Objective: KaaS to improve the production quality. Method: Interviews, focus group discussions, validation by 22 experts, content analysis and systemic review of $48.8 \%$ of the bibliography of 332 references. Results: A conceptual advance was achieved in the design of the prototype of the final architecture for the proposed technological solution. Conclusions: The digitization of the company from its business architecture facilitates the democratization of knowledge evidenced in the case study.
Due to social, economic, and environmental problems, the need arises to develop what is known as a Smart City allowing to improve the quality of people’s lives together with the application of emerging technologies such as Blockchain and providing improvements in different areas such as medical care, intelligent transport, or the supply chain management. The investigation analyses the association between Blockchain and Smart Cities using a Bibliometric Analysis, collected data from 384 articles published between 2018 and October 2022 with the topics ‘Blockchain’ and ‘Smart Cities’ from the Web of Science database. It has executed the VOSviewer program to appreciate the Bibliometric Analysis. The work has identified six research trends related with these fields.
The scoping review reported by this article aimed to analyze and synthesize state-of-the-art studies focused on the integration of cyber resilience in the implementation of smart cities. An electronic search was conducted, and 11 studies were included in this review after the selection process. According to the findings, cyber resilience represents a gap of the current research related to smart cities and, therefore, additional efforts are required to guarantee that smart cities are resilient to challenging events such as cyber-attacks or natural disasters.
