Examinando por Materia "Data analysis"

Mostrando 1 - 6 de 6
  • No hay miniatura disponible
    Ítem
    Adaptive learning objects in the context of eco-connectivist communities using learning analytics
    (Elsevier BV, 2019-11-01) Diego, Mosquera; Carlos, Guevara; Jose, Aguilar; Diego, Mosquera; Carlos, Guevara; Jose, Aguilar; Universidad EAFIT. Departamento de Ingeniería de Sistemas; I+D+I en Tecnologías de la Información y las Comunicaciones
    Eco-connectivist communities are groups of individuals with similar characteristics, which emerge in a connectivist learning process within a knowledge ecology. ARMAGAeco-c is a reflexive and autonomic middleware for the management and optimization of eco-connectivist knowledge ecologies using description, prediction and prescription models. Adaptive Learning Objects are autonomic components that seek to personalize Learning Objects according to certain contextual information, such as learning styles of the learner's, technological restrictions, among other aspects. MALO is a system that allows the management of Adaptive Learning Objects. One of the main challenges of the connectivist learning process is the adaptation of the educational context to the student needs. One of them is the learning objects. For this reason, this work has two objectives, specifying a data analytics task to determine the learning style of a student in an eco-connectivist community and, adapting instances of Adaptive Learning Objects using the learning styles of the students in the communities. We use graph theory to identify the referential member of each eco-connectivist community, and a learning paradigm detection algorithm to identify the set of activities, strategies, and tools that Adaptive Learning Objects instances should have, according to the learning style of the referential member. To test our approach, a case study is presented, which demonstrates the validity of our approach.
  • No hay miniatura disponible
    Ítem
  • No hay miniatura disponible
    Ítem
    Aspectos a tener en cuenta para una legislación sobre el tratamiento de datos de niños niñas y adolescentes en redes sociales : un análisis desde el derecho comparado
    (Universidad EAFIT, 2023) Quintero Toro, Jorge Daniel; Higuita Olaya, Daniel
    Digital social networks and access to technology have transcended the physical barriers under which information was observed. People is increasingly connected and has higher access to information, what is also increasing the risks they face on-the-grid. This is partly because of the way in which social networks are using artificial intelligence to collect data generated from user’s interactions with the network. Such information is being used to profile users for future advertising or simply to keep them connected for longer periods, so the artificial intelligence can learn more from the way the user interacts with the network. Colombian law prohibits the collection of personal data of children and adolescents. However, this prohibition has no effect, because there is easy access to digital social networks. This implies risks that can be mitigated with more efficient laws that are consistent with current reality. This entails risks for children and adolescents who, depending on their mental maturity and decision-making capacity, may or may not be able to assume these risks. Some of the risks to which children and adolescents can be subjected are the radicalization of thoughts, limiting the free development of the personality, habeas data, extortion or sexual crimes. This research becomes relevant because there are no studies about this, which implies many social risks by directly influencing the future of society such as children and adolescents.
  • No hay miniatura disponible
    Ítem
    An Automatic Merge Technique to Improve the Clustering Quality Performed by LAMDA
    (Institute of Electrical and Electronics Engineers Inc., 2020-01-01) Morales, Luis; Aguilar, Jose; Morales, Luis; Aguilar, Jose; Universidad EAFIT. Departamento de Ingeniería de Sistemas; I+D+I en Tecnologías de la Información y las Comunicaciones
    Clustering is a research challenge focused on discovering knowledge from data samples whose goal is to build good quality partitions. In this paper is proposed an approach based on LAMDA (Learning Algorithm for Multivariable Data Analysis), whose most important features are: a) it is a non-iterative fuzzy algorithm that can work with online data streams, b) it does not require the number of clusters, c) it can generate new partitions with objects that do not have enough similarity with the preexisting clusters (incremental-learning). However, in some applications, the number of created partitions does not correspond with the number of desired clusters, which can be excessive or impractical for the expert. Therefore, our contribution is the formalization of an automatic merge technique to update the cluster partition performed by LAMDA to improve the quality of the clusters, and a new methodology to compute the Marginal Adequacy Degree that enhances the individual-cluster assignment. The proposal, called LAMDA-RD, is applied to several benchmarks, comparing the results against the original LAMDA and other clustering algorithms, to evaluate the performance based on different metrics. Finally, LAMDA-RD is validated in a real case study related to the identification of production states in a gas-lift well, with data stream. The results have shown that LAMDA-RD achieves a competitive performance with respect to the other well-known algorithms, especially in unbalanced benchmarks and benchmarks with an overlapping of around 9%. In these cases, our algorithm is the best, reaching a Rand Index (RI) >98%. Besides, it is consistently among the best for all metrics considered (Silhouette coefficient, modification of the Silhouette coefficient, WB-index, Performance Coefficient, among others) in all case studies analyzed in this paper. Finally, in the real case study, it is better in all the metrics.
  • No hay miniatura disponible
    Ítem
    El conocimiento es poder entendiendo la correlación entre la información recolectada a través de un perfil de Instagram y el consumo de contenido de forma periódica y repetitiva en dicha plataforma
    (Universidad EAFIT, 2020) Rieder Monsalve, Paulina; Toro Valencia, Jose Alberto
    Life in the XXI century happens trought Internet, and more specifically, trought social media. This plataforms attract users attention by costumization of content, wich is constructed from the knowledge extracted in the data mining process thanks to the personal data that every user yields with the use of the service. The purpose of personalization is the augmentation of time spent online thus, the amount of personal data given. This data is the fundamental asset to digital economy. Due to the wide knowledge derived from data mining and the profound social incertion that social media has, users get trapped onto a asimetric power relationship with the companies that provide this service wich directly impacts rights such as privacy. Even though privacy has international and national regulations, its guideline is not appropiate to face the current challenges.
  • No hay miniatura disponible
    Ítem
    Fuzzy electre model for the characterisation of aeronautical operational risks in the approach and landing phase
    (Universidad EAFIT, 2023) Leal Cabra, Estefanía del Pilar; Peña, Juan Alejandro
    One of the significant challenges facing the aviation sector is the management of risks arising from its flight operations, especially in the approach and landing phases, where pilot experience and training are of great importance and where the most significant incidents for air safety occur. Therefore, this paper proposes a model inspired by the structure of a Fuzzy ELECTRE model for managing the operational risks that arise in the approach and landing phases that can lead to safety events. Thanks to the analysis of the literature collected, the management criteria and risk parameters to be taken into account for these two flight phases were shown following air safety manuals such as the International Civil Aviation Organization (ICAO) manual, and where the data obtained was obtained qualitatively thanks to the implementation of surveys with expert pilots, whose information served as the primary input for the characterisation of risks. Following the structure of the proposed model, five (5) reference risk scenarios management were constructed using the previous information, and an analysis of the dominance and discrepancy of a risk scenario vs. the previously established reference scenarios was carried out. Finally, it can be concluded that the proposed model allowed the quantitative-qualitative characterisation for managing the most relevant risks in the approach and landing phases, integrating the expertise of experts in this area.