In Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy
will defend his proposal
Geographically Distributed Orchestration System for Collection of Training Data: Enabling a Supervised Learning Engine for Service Identification
AbstractThe internet has become a backbone in terms of supporting people’s needs in so many aspects of everyday life. While spending time on the internet, each person generates a unique footprint which could lead to identifying each user. Identifying the users by tracing their internet footprints shows a dual relationship between users and the services that are being used. That being said, in order to identify users, it is essential to identify the services. For such complex identification procedures, data science is one of the primary resources from which to get support. Even though data science can provide help on so many levels, producing a well-formed data set is challenging. This thesis focuses on the collecting of high-grade data sets which will also be used on determining service behavior identification. The data collection system created for this thesis consists of geographically-dispersed data gathering points. These data gathering points are configurable to generate particular isolated service data which are identical to real user footprints. Preliminary work includes implementation of data collection methods and comprehensive data exploration.
Date: Monday, May 13, 2019
Time: 10:00 - 11:00 AM
Place: T2, Room 200
Advisor: Dr. Deniz Gurkan
Faculty, students, and the general public are invited.