- Neural Networks and Applications
- Machine Learning and ELM
- Face and Expression Recognition
- Metaheuristic Optimization Algorithms Research
- Online Learning and Analytics
- Domain Adaptation and Few-Shot Learning
- Imbalanced Data Classification Techniques
- Online and Blended Learning
- Evolutionary Algorithms and Applications
- Financial Distress and Bankruptcy Prediction
- Complex Systems and Time Series Analysis
- Educational Innovations and Technology
- E-Learning and Knowledge Management
- Job Satisfaction and Organizational Behavior
- Machine Learning and Data Classification
- Mineral Processing and Grinding
- COVID-19 Pandemic Impacts
- Risk and Portfolio Optimization
- Stock Market Forecasting Methods
- COVID-19 epidemiological studies
- Ecosystem dynamics and resilience
- Intimate Partner and Family Violence
- Advanced Multi-Objective Optimization Algorithms
- Technology-Enhanced Education Studies
- Grey System Theory Applications
Universidad de Málaga
2022-2025
Laboratoire de Réactivité et Chimie des Solides
2024
Université de Picardie Jules Verne
2024
Centre National de la Recherche Scientifique
2024
Réseau sur le Stockage Electrochimique de l'énergie
2024
Universidad Loyola Andalucía
2014-2022
IVI Sevilla Clinic
2020
Universidad Loyola
2014-2018
Loyola University Chicago
2015
European Space Research and Technology Centre
2012-2014
Ordinal regression problems are those machine learning where the objective is to classify patterns using a categorical scale which shows natural order between labels. Many real-world applications present this labelling structure and that has increased number of methods algorithms developed over last years in field. Although ordinal can be faced standard nominal classification techniques, there several specifically benefit from ordering information. Therefore, paper aimed at reviewing state...
Artificial neural networks (ANNs) have traditionally been seen as black-box models, because, although they are able to find "hidden" relations between inputs and outputs with a high approximation capacity, their structure seldom provides any insights on the of functions being approximated. Several research papers tried debunk nature ANNs, since it limits potential use ANNs in many areas. This paper is framed this context proposes methodology determine individual collective effects input...
This article explores the quality of online learning experience based on Sloan-C framework and Online Learning Consortium's (OLC) scorecard. The OLC index has been implemented to evaluate in programs from different perspectives. Despite this, opinions learners are ignored, it is built using feedback experts panelists while ignoring factors that teachers consider important during their lectures. We propose an alternative way measuring by analyzing satisfaction students' perceptions. 11...
In this paper, the well known stagewise additive modeling using a multiclass exponential (SAMME) boosting algorithm is extended to address problems where there exists natural order in targets cost-sensitive approach. The proposed ensemble model uses an extreme learning machine (ELM) as base classifier (with Gaussian kernel and additional regularization parameter). closed form of derived weighted least squares problem provided, it employed estimate analytically parameters connecting hidden...
ABSTRACT The alignment between organizational and employee human values is critical in institutions that base their management on values. This research aims to link Schwartz's 10 workplace authenticity, determining prevalent organizations. In tandem, nonprofit organizations are intricately intertwined with the held by members, forming bedrock of identity. To achieve this goal, an in‐depth analysis conducted three faith‐based study proposes a hybrid model, merging genetic algorithm linear...
Transient instability is considered the most severe form of in power systems with grave socioeconomic repercussions if not prevented. Conventional methods, such as time domain simulations and direct methods impose limitations to fast on-line transient stability assessment modern systems. The development phasor measurement units paved way for by means artificial intelligence pattern recognition classification. Many classification algorithms have been reported literature assessing stability....
Ordinal regression (OR) is an important branch of supervised learning in between the multiclass classification and regression. In this paper, traditional scheme neural network adapted to learn ordinal ranks. The model proposed imposes monotonicity constraints on weights connecting hidden layer with output layer. To do so, are transcribed using padding variables. This reformulation leads so-called inequality constrained least squares (ICLS) problem. Its numerical solution can be obtained by...
The Job Demand-Control and Demand-Control-Support (JDCS) models constitute the theoretical approaches used to analyze relationship between characteristics of labor occupational health. Few studies have investigated main effects multiplicative model in relation perceived health professional accountants. Accountants are subject various types pressure performing their work; this influences and, ultimately, ability perform a job well. objective study is investigate demands on 739 accountants, as...
The topic of online instructors’ roles has been interest to the educational community since late twentieth century. In previous studies, identification was done using a top-down (deductive) approach. This study applied bottom-up (inductive) procedure examine not only instructors from student perspective, but also how well these are implemented in practice. first stage, were defined factor analysis on sample 925 students. A questionnaire created after an extensive literature review and...
In this paper, two neural network threshold ensemble models are proposed for ordinal regression problems. For the first method, thresholds fixed a priori and not modified during training. The second one considers of each member as free parameters, allowing their modification training process. This is achieved through reformulation these tunable thresholds, which avoids constraints they must fulfill problem. During training, diversity exists in different projections generated by taken into...
The current European debt crisis has drawn considerable attention to credit-rating agencies' news about sovereign ratings. From a technical point of view, credit rating constitutes typical ordinal regression problem because agencies generally present scale risk composed several categories. This fact motivated the use an approach address in this paper. Therefore, ranking different classes will be taken into account for design classifier. To do so, novel model is introduced order replicate...
<p class="3">This paper determines which instructional roles and outputs are important in the 21<sup>st</sup> century from perspective of students asynchronous learning environments. This research work uses a literature review, in-depth interviews with experts, pilot study to define instructors’ outputs. Following this, determined by using quantitative methodology (in sample 925 students). To our knowledge, remaining works on this topic identify online instructors'...
This study examines success factors in online learning from the instructors' perspective. Academic comprises not only student satisfaction and good grades, but also perception of knowledge transfer. A systemic model inputs–process–outputs was used. total 322 teachers four different universities countries were used to attainment. Findings suggest that: (i) instructors University Peking Autonomous Popular State Puebla reported learner as most important for students on courses, (ii) New Mexico...