Advanced techniques for efficient privacy preserving data mining software
Προηγμένες τεχνικές για αποδοτικό λογισμικό εξόρυξης δεδομένων διατήρησης απορρήτου
View/
Keywords
Privacy ; Data mining ; Transactional data bases ; Knowledge graphs ; Natural language processing ; Machine learning ; Deep learning ; Frequent Itemset Hiding Problem ; Knowledge hidingAbstract
Privacy preservation is paramount in the era of data-driven decision-making, particularly in domains such as data mining and legal document analysis. This work investigates the intersection of privacy-preserving data mining techniques and Natural Language Processing (NLP) to address concerns related to privacy risks and data confidentiality. The research focuses on developing innovative approaches and methodologies to safeguard individual privacy while extracting meaningful insights. In particular, it focuses on the problems of Frequent Itemset Hiding and NER for low-resource languages or domains. Leveraging advanced data mining algorithms and techniques, the study explores strategies for managing privacy challenges in data sharing. Through comprehensive experimentation and analysis, this thesis contributes novel insights and practical solutions for enhancing privacy, while considering data quality and strives to find solutions that are suitable for real-world applications. The findings offer valuable contributions to both academic research and practical applications in the privacy domain.
Department
Σχολή Τεχνολογιών Πληροφορικής και Επικοινωνιών. Τμήμα ΠληροφορικήςNumber of pages
170Language
EnglishCollections
Except where otherwise noted, this item's license is described as
Αναφορά Δημιουργού-Μη Εμπορική Χρήση-Όχι Παράγωγα Έργα 3.0 Ελλάδα
Αναφορά Δημιουργού-Μη Εμπορική Χρήση-Όχι Παράγωγα Έργα 3.0 Ελλάδα