Distributed Privacy-Preserving exact - area under the curve; a novel method to compute the exact global AUC without revealing individual sensitive input data. It utilizes a combination of Paillier, symmetric, and asymmetric encryption with perfect randomized encoding to compute the exact measurement even with tie conditions.
Run pip install -r requirements.txt to ensure all requirements are fulfilled.
Sequence data of HIV-1 strains and the corresponding coreceptor binding of 10462 subjects was used. The raw data is from
www.hiv.lanl.gov and processed with /showcase/data/data_preparing.py. A binary classifier was trained and the performance
is evaluated against the commonly standard sklearn AUC library. Over 10 runs, the performance difference is within e-16 which
is due to floating type conversions in python. Furthermore, a evaluation of the runtime regarding random number generation
and flag data generation was done.
Two different experiments are used to measure the runtime with synthetic data. To generate sample data, specify the number of stations and subjects. In the following experiments 10-60% of flag subjects are added randomly.
The performance is evaluated against the commonly standard sklearn AUC library.
In experiment 1 the number of samples are identical, but the number of input parties varies. The execution time is calculated absolute with each step and party for 10 runs. Therefore, the total execution time differs with the number of input parties.
In experiment 2 the number of samples are increasing with 3 stations. The total execution increases with the number of samples to encrypt and decrypt.
