Bilkent University
Department of Computer Engineering


Predicting Optimal Facility Location Using Synthetic Data Generation


Sanem Elbaşı
MS Student
Computer Engineering Department
Bilkent University

Location analytics is the process or the ability to gain insight from the location data. Businesses use location analytics in many ways such as finding the optimal place to locate a new facility, identifying the performances of stores. However, businesses do not own the location data of its customers most of the time. In this work, we consider the problem of finding the optimal location for the new store of a business while optimizing an objective function when the customer location information is not present. We propose using synthetic data when the original data is not available. Synthetic data can be generated by businesses using neighborhood-based data generation or by a server, which owns the original data but cannot publish this data as it is, to anonymize the data before publishing it. For server-side approach, we propose grid-based and clustering-based data generation methods, which satisfies differential privacy and high utility.


DATE: 02 April, 2018, Monday, CS590 & CS690 presentations begin at @ 15:40