A5048207145- Face and Expression Recognition
- Neural Networks and Applications
- Machine Learning and Data Classification
- Gene expression and cancer classification
- Anomaly Detection Techniques and Applications
- Evolutionary Algorithms and Applications
- Data Mining Algorithms and Applications
- Data Stream Mining Techniques
- Network Security and Intrusion Detection
- Imbalanced Data Classification Techniques
- Blind Source Separation Techniques
- Healthcare Technology and Patient Monitoring
- Fault Detection and Control Systems
- Machine Learning in Bioinformatics
- Neonatal and fetal brain pathology
- Advanced Image and Video Retrieval Techniques
- Control Systems and Identification
- Fuzzy Logic and Control Systems
- Machine Learning in Healthcare
- Machine Learning and ELM
- Non-Invasive Vital Sign Monitoring
- AI-based Problem Solving and Planning
- EEG and Brain-Computer Interfaces
- Image Retrieval and Classification Techniques
- Artificial Intelligence in Healthcare
Universidade da Coruña
2015-2024
Polytechnic Institute of Porto
2023
Tilburg University
2019
Augusta University
1989-2005
Universidade de Santiago de Compostela
1988-2005
With the advent of large-scale problems, feature selection has become a fundamental preprocessing step to reduce input dimensionality. The minimum-redundancy-maximum-relevance (mRMR) selector is considered one most relevant methods for dimensionality reduction due its high accuracy. However, it computationally expensive technique, sharply affected by number features. This paper presents fast-mRMR, an extension mRMR, which tries overcome this computational burden. Associated with we include...
Discretization of numerical data is one the most influential preprocessing tasks in knowledge discovery and mining. The purpose attribute discretization to find concise representations as categories which are adequate for learning task retaining much information original continuous possible. In this article, we present an updated overview techniques conjunction with a complete taxonomy leading discretizers. Despite great impact technique, few elementary approaches have been developed...
With the advent of extremely high dimensional datasets, dimensionality reduction techniques are becoming mandatory. Of many available, feature selection (FS) is growing interest for its ability to identify both relevant features and frequently repeated instances in huge datasets. We aim demonstrate that standard FS methods can be parallelized big data platforms like Apache Spark so as boost performance accuracy. propose a distributed implementation generic framework includes broad group...