Quantifying Temporal Privacy Leakage in Continuous Event Data Publishing
Project description
Introduction
This project implements the quantification of privacy leakage for differential privacy mechanisms in continuous event data publishing.
Python package
The implementation has been published as a standard Python package. Use the following command to install the corresponding Python package:
pip install pm-cedp-qdp
Usage
from pm_cedp_qdp.qdp import QDP
if __name__ == '__main__':
log_name = "Sepsis Cases.xes"
max_num_release = 5
state_window = 200 # A large number will consider the entire prefix/suffix of traces
state_direction = "backward" # Backward (prefix) or forward (suffix)
event_percentage = 0.5 # The percentage of events included in the first release
recursive = True # This will continue quantifying releases until there is no incomplete trace or reaching max_num_release. Otherwise, only one release is quantified.
only_complete_traces = False # If one wants to only consider the complete traces for generating temporal correlations.
epsilon = 0.01 # The privacy parameter of DP mechanism
window_size = 2 # How many new events per trace are released at each publish.
certain_release = True # If this is true, the number of new events per variant in each new release is equal to window_size, otherwise it is a random number range from 0 to window_size
explore_depth = window_size # The depth of exploring the transitions system.
export_csv = log_name[:-4] + "_" + str(state_window) + "_" + state_direction + "_" + str(certain_release) + str(window_size) + ".csv"
qdp = QDP()
release_index, BPL_list, FPL_list, TPL_list = qdp.apply(log_name, epsilon, export_csv, recursive=recursive,
only_complete_traces=only_complete_traces, state_window=state_window,
state_direction=state_direction,
explore_depth = explore_depth,
window_size= window_size,
certain_release= certain_release,
event_percentage= event_percentage,
max_num_release = max_num_release)
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
pm_cedp_qdp-0.1.2.tar.gz
(20.9 kB
view details)
Built Distribution
File details
Details for the file pm_cedp_qdp-0.1.2.tar.gz
.
File metadata
- Download URL: pm_cedp_qdp-0.1.2.tar.gz
- Upload date:
- Size: 20.9 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/3.4.2 importlib_metadata/4.8.1 pkginfo/1.4.2 requests/2.26.0 requests-toolbelt/0.9.1 tqdm/4.62.3 CPython/3.6.5
File hashes
Algorithm | Hash digest | |
---|---|---|
SHA256 | eaaeb95405f14a7cb4599018d955b72080a4cd60947c5228be8fcec155384f83 |
|
MD5 | a032de9fcf55aefd2cdfa5ab0090cd16 |
|
BLAKE2b-256 | 35e92b8cf8095368b5913ea320241d0a1c012c16218740449ef2d4501cbcc37f |
File details
Details for the file pm_cedp_qdp-0.1.2-py3-none-any.whl
.
File metadata
- Download URL: pm_cedp_qdp-0.1.2-py3-none-any.whl
- Upload date:
- Size: 21.2 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/3.4.2 importlib_metadata/4.8.1 pkginfo/1.4.2 requests/2.26.0 requests-toolbelt/0.9.1 tqdm/4.62.3 CPython/3.6.5
File hashes
Algorithm | Hash digest | |
---|---|---|
SHA256 | a1434aede4d33f0c4a3f8ec2e245927f03e744aa2e51233e9ae89149ca108c11 |
|
MD5 | b14ff3d4edf038d3a6c43c13210223ba |
|
BLAKE2b-256 | 8ecdd322756cc803af180cc7e5351b468a37e0533fc5421100bbe818f1b34b83 |