A5035687300- Reinforcement Learning in Robotics
- Bayesian Modeling and Causal Inference
- Advanced Image and Video Retrieval Techniques
- Robotics and Sensor-Based Localization
- Evolutionary Algorithms and Applications
- Robotic Path Planning Algorithms
- Text and Document Classification Technologies
- AI-based Problem Solving and Planning
- Machine Learning and Data Classification
- Image Retrieval and Classification Techniques
- Metaheuristic Optimization Algorithms Research
- Semantic Web and Ontologies
- Anomaly Detection Techniques and Applications
- Machine Learning in Bioinformatics
- Data Stream Mining Techniques
- Advanced Multi-Objective Optimization Algorithms
- Fault Detection and Control Systems
- Modular Robots and Swarm Intelligence
- Face and Expression Recognition
- Data Mining Algorithms and Applications
- Robot Manipulation and Learning
- Human Pose and Action Recognition
- Rough Sets and Fuzzy Logic
- Domain Adaptation and Few-Shot Learning
- Logic, Reasoning, and Knowledge
National Institute of Astrophysics, Optics and Electronics
2016-2025
Iraqi University
2025
Universitas Tama Jagakarsa
2024
Presidency University
2024
Mathematics Research Center
2020-2022
First Pavlov State Medical University of St. Petersburg
2021
Mansoura University
2018
Mexican Academy of Sciences
2013
Instituto de Investigaciones Eléctricas
1990-2009
Tecnológico de Monterrey
1997-2006
In multidimensional classification the goal is to assign an instance a set of different classes. This task normally addressed either by defining compound class variable with all possible combinations classes (label power-set methods, LPMs) or building independent classifiers for each (binary-relevance BRMs). However, LPMs do not scale well and BRMs ignore dependency relations between We introduce method chaining binary Bayesian that combines strengths classifier chains networks...
Wearable health devices have emerged as a transformative force in healthcare, bridging the gap between digital technology and personalized medicine. These enable users to monitor various metrics, from heart rate activity levels sleep quality stress, contributing proactive approach management. Concurrently, data analytics has advanced utility of these by offering actionable insights, enhancing clinical decision-making, driving care. This paper examines trends wearable technology, role...
The ability to recognize people is a key element for improving human-robot interaction in service robots.There are many approaches face recognition; however, these assume unrealistic conditions robot, like having an image with centered under controlled illumination.We have developed novel recognition system so that mobile robot can learn new faces and them real-time realistic indoor environments.It able on-line based on single frame, which later used the person even different environmental...
Imbalanced data sets in the class distribution is common to many real world applications. As classifiers tend degrade their performance over minority class, several approaches have been proposed deal with this problem. In paper, we propose two new cluster-based oversampling methods, SOI-C and SOI-CJ. The methods create clusters from instances generate synthetic inside those clusters. contrast other avoid creating majority regions. They are more robust noisy examples (the number of generated...
Stress assessment is a complex issue and numerous studies have examined factors that influence stress in working environments. Research shown monitoring individuals' behaviour parameters during daily life can also help assess levels. In this study, we examine of work-related using features derived from sensors smartphones. particular, use information physical activity levels, location, social-interactions, social-activity, application usage days. Our study included 30 employees chosen two...