A5051554470- Privacy-Preserving Technologies in Data
- Spam and Phishing Detection
- Stochastic Gradient Optimization Techniques
- Biomedical Text Mining and Ontologies
- Blockchain Technology Applications and Security
- Machine Learning in Healthcare
- Mobile Crowdsensing and Crowdsourcing
- Caching and Content Delivery
- Web Data Mining and Analysis
- Distributed and Parallel Computing Systems
- Advanced Text Analysis Techniques
- Topic Modeling
- Natural Language Processing Techniques
- Scientific Computing and Data Management
- Advanced Data Storage Technologies
- Internet Traffic Analysis and Secure E-voting
- Network Security and Intrusion Detection
- Sentiment Analysis and Opinion Mining
- Cryptography and Data Security
- Traffic Prediction and Management Techniques
Sofia University "St. Kliment Ohridski"
2011-2024
Institute of Mathematics and Informatics
2023
Groupe d'Analyse et de Théorie Economique Lyon St Etienne
2023
The task of automatic diagnosis encoding into standard medical classifications and ontologies is great importance in medicine -both to support the daily tasks physicians preparation reporting clinical documentation, for processing reports.In this paper, we investigate application performance different deep learning transformers ICD-10 texts Bulgarian.The comparative analysis attempts find which approach more efficient be used finetuning pre-trained BERT family transformer deal with a...
The World Wide Web provides continuous sources of information with similar semantic structure like news feeds, user reviews and comments on various topics. These are essential for the goal online opinion mining. paper proposes a computationally efficient algorithm structured extraction from web pages. relies combination analysis data natural language processing text content. It maps HTML pages containing news, or to custom designed RSS feed structure. Such usually includes textual opinions,...
Nowadays, data-sharing ecosystems are crucial for unlocking and realizing the maximum potential of data. Data spaces an emergent concept that helps to overcome some challenges related data sharing supports creation innovative solutions in a trustful mutually beneficial manner. This paper shows how competing companies mobility domain can collaborate toward optimizing performance traffic prediction algorithm through implementing federated machine learning space. The proposed method avoids...
Federated learning (FL) is a decentralized machine approach where independent learners process data privately. Its goal to create robust and accurate model by aggregating retraining local models over multiple rounds. However, FL faces challenges regarding heterogeneity aggregation effectiveness. In order simulate real-world data, researchers use methods for partitioning that transform dataset designated centralized into group of sub-datasets suitable distributed with different heterogeneity....
Federated learning (FL) is a decentralized machine approach where independent learners process data privately. Its goal to create robust and accurate model by aggregating retraining local models over multiple rounds. However, FL faces challenges regarding heterogeneity aggregation effectiveness. In order simulate real-world data, researchers use methods for partitioning that transform dataset designated centralized into group of sub-datasets suitable distributed with different heterogeneity....
Federated Learning is a machine learning technique where independent devices (clients) cooperatively train model by working on decentralized training data. A fundamental challenge in federated the client aggregation. The goal to combine and preserve knowledge that each has acquired during its local phase generate an aggregated with superior performance than any of clients. Most current state-of-the-art algorithms for aggregation still rely simple weight parameter averaging. Though effective,...