A5019105782- Advanced Malware Detection Techniques
- Context-Aware Activity Recognition Systems
- Bullying, Victimization, and Aggression
- Hate Speech and Cyberbullying Detection
- User Authentication and Security Systems
- Biometric Identification and Security
- Network Security and Intrusion Detection
- Impact of Technology on Adolescents
- Advanced Bandit Algorithms Research
- Consumer Retail Behavior Studies
- Emotion and Mood Recognition
- Generative Adversarial Networks and Image Synthesis
- Human Pose and Action Recognition
- Age of Information Optimization
- IoT-based Smart Home Systems
- Technology Use by Older Adults
- Digital Media Forensic Detection
- Technology Adoption and User Behaviour
- Solar Radiation and Photovoltaics
- Greenhouse Technology and Climate Control
- Forensic Fingerprint Detection Methods
- Biomedical Text Mining and Ontologies
- Human Mobility and Location-Based Analysis
- Child Development and Digital Technology
- Smart Agriculture and AI
University of Bari Aldo Moro
2020-2024
Human Activity Recognition (HAR) is an essential area of research related to the ability smartphones retrieve information through embedded sensors and recognize activity that humans are performing. Researchers have recognized people's activities by processing data received from with Machine Learning Models. This work intended be a hands-on survey practical's tables capable guiding reader used in modern highly cited developed machine learning models perform human recognition. Several papers...
The smartphone is an excellent source of data; it possible to extrapolate sensor values and, through Machine Learning approaches, perform anomaly detection analysis characterized by human behavior. This work exploits Human Activity Recognition (HAR) models and techniques identify activity performed while filling out a questionnaire via application, which aims classify users as Bullying, Cyberbullying, Victims Cyberbullying. purpose the discuss new methodology that combines final label...
Bullying and cyberbullying are harmful social phenomena that involve the intentional, repeated use of power to intimidate or harm others. The ramifications these actions felt not just at individual level but also pervasively throughout society, necessitating immediate attention practical solutions. BullyBuster project pioneers a multi-disciplinary approach, integrating artificial intelligence (AI) techniques with psychological models comprehensively understand combat issues. In particular,...
Abstract This study establishes a correlation between computer science and psychology, specifically focusing on the incorporation of smartphone sensors users' personality index. A limited number state-of-the-art approaches have considered these factors, while no existing dataset currently encompasses this correlation. In study, an Android application was developed to implement questionnaire bullying cyberbullying, using predict Personal Index. Sensor data are collected in “ UNIBA HAR Dataset...
The security of modern smartphones is related to the combination Continuous Authentication approaches, Touch events, and Human Activities. approaches Authentication, Events, Activities are silent user but a great source data for Machine Learning Algorithms. This work aims develop method continuous authentication while sitting scrolling documents on smartphone. Events Smartphone Sensor Features (from well-known H-MOG Dataset) were used with addition, each sensor, feature called Signal Vector...
Today's Web aims to increase engagement with the user, his or her emotions and behaviors. Contemplating emotional sphere is potentially very useful when it comes sites: understanding how user feels can help service provider understand needs improve he she offers. This project devise a behavioral feature extraction system create an interactive that provide personalized experience maximize satisfaction. The contemplates analysis techniques using Machine Learning, Recommender System Emotion...
Mobile devices are becoming increasingly popular. The widespread use exposes individuals to unintentionally sharing sensitive information that could allow direct access the mobile device. This paper implements and tests a user verification system based on touch dynamics biometrics while typing Personal Identification Number (PIN). proposed framework includes several features feature selection. classification problem has been considered in terms of one-class binary classifiers. approach can...
This work considers the Internet of Things (IoT) and machine learning (ML) applied to agricultural sector within a real-working scenario. More specifically, aim is punctually forecast two most important meteorological parameters (solar radiation rainfall) determine amount water needed by specific plantation under different contour conditions. Three state-of-the-art ML approaches, coupled with boosting techniques, have been adopted compared obtain hourly forecasting. Real-working conditions...
In this paper, an algorithm, for in-parallel, greedy experience generator (briefly IPE, Parallel Experiences), has been crafted, and added to the Double Deep Q-Learning algorithm. The algorithm aims perturbs weights of online network, as results, trying recover from perturbed weights, escapes local minima. DDQN with IPE takes about double time previous compute, but even if it slows down learning rate in terms wall clock time, solution converges faster number epochs.