Aristotle University Of Thessaloniki (AUTH)
Ανάπτυξη υπηρεσιών και χαρακτηρισμός χρηστών επικοινωνιακής & δικτυακής τεχνολογίας
Abstract
dc:descriptionThe main task of the dissertation is the modeling of user behavior in a telecommunications network in order to detect fraudulent activities. User behavior characterization is achieved by means of appropriate user profiles. These can be either one- or multi-dimensional structures. One-dimensional user activity representations are actually time-series of a network usage measure, e.g. calls per day or bulk of downloaded data. Multi-dimensional profiles may use more usage characteristics such as calls per day, corresponding duration, cost, hour-of-day, etc. User profiles are analyzed through artificial intelligence, statistical and data mining methods. The dissertation uses both a dofferential and an absolute approach to the problem. Problem heuristics as well as all experimentation and conclusions are based on real world
Degree
thesis:*- Grantor dc:publisher
- Aristotle University Of Thessaloniki (AUTH)
- Year dc:date
- 2006
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Hilas, Constantinos
- Χειλάς, Κωνσταντίνος
Subjects
dc:subject × 22- Ανίχνευση απάτης
- Χαρακτηρισμός και πορτραίτα χρηστών
- Τηλεπικοινωνίες
- Δίκτυα
- Εξόρυξη δεδομένων
- Ανάλυση χρονοσειρών
- Ανάλυση παρεμβάσεων
- Έμπειρα συστήματα στις τηλεπικοινωνίες
- Νοιμικά και οικονομικά θέματα στις τηλεπικοινωνίες
- Fraud detection
- User profiling & characterization
- Telecommunications
- Networking
- Data mining
- Time-series analysis
- Intervention analysis
- Expert systems in telecommunications
- Legal and economic aspects of telecommunications
- Φυσικές Επιστήμες
- Φυσική
- Natural Sciences
- Physical Sciences
Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/19288