| KEYWORD |
Observability for network services - Time series analysis for anomaly detection and pattern recognition
Thesis in external company
keywords ANOMALY DETECTION, COMPUTER AIDED DIAGNOSIS, NETWORK MONITORING, PATTERN RECOGNITION
Reference persons LUCA VASSIO
External reference persons Giorgio Bernardi - INRETE - giorgio.bernardi@inrete.it - http://www.inrete.it
Research Groups Telecommunication Networks Group
Thesis type SPERIMENTALE
Description The increasingly pervasive presence of wired and wireless connectivity services requires the development of methodologies and metrics for functional validation and performance measurement in the networkds. The company INRETE, in carrying out its Observability activities of network services aimed at the enterprise world, collects a large number of indicators on the various elementary components enabling network services.
Extracting information on trends and operating patterns from a large number of time series is not a trivial task, especially if you want the type of information extracted to have explicability characteristics that allow it to be immediately understood by a human operator.
The thesis proposal therefore focuses on the ability to extract trends and patterns in order to enable dialogues with humans in a "computer aided" perspective and not full automation/human replacement.
Required skills Computer networks,
Data mining,
Machine learning,
Programming (Python)
Notes The thesis is developed in the company INRETE -- http://www.inrete.it
You can also contact - giorgio.bernardi@inrete.it
Deadline 06/11/2025
PROPONI LA TUA CANDIDATURA