anonypy's Introduction

AnonyPy

Anonymization library for python. AnonyPy provides following privacy preserving techniques for the anonymization.

K Anonymity
L Diversity
T Closeness

The Anonymization method

Anonymization method aims at making the individual record be indistinguishable among a group record by using techniques of generalization and suppression.
Turning a dataset into a k-anonymous (and possibly l-diverse or t-close) dataset is a complex problem, and finding the optimal partition into k-anonymous groups is an NP-hard problem.
AnonyPy uses "Mondrian" algorithm to partition the original data into smaller and smaller groups
The algorithm assumes that we have converted all attributes into numerical or categorical values and that we are able to measure the “span” of a given attribute Xi.

Install

$ pip install anonypy

Usage

import anonypy
import pandas as pd

data = [
    [6, "1", "test1", "x", 20],
    [6, "1", "test1", "x", 30],
    [8, "2", "test2", "x", 50],
    [8, "2", "test3", "w", 45],
    [8, "1", "test2", "y", 35],
    [4, "2", "test3", "y", 20],
    [4, "1", "test3", "y", 20],
    [2, "1", "test3", "z", 22],
    [2, "2", "test3", "y", 32],
]

columns = ["col1", "col2", "col3", "col4", "col5"]
categorical = set(("col2", "col3", "col4"))

def main():
    df = pd.DataFrame(data=data, columns=columns)

    for name in categorical:
        df[name] = df[name].astype("category")

    feature_columns = ["col1", "col2", "col3"]
    sensitive_column = "col4"

    p = anonypy.Preserver(df, feature_columns, sensitive_column)
    rows = p.anonymize_k_anonymity(k=2)

    dfn = pd.DataFrame(rows)
    print(dfn)

Original data

   col1 col2   col3 col4  col5
0     6    1  test1    x    20
1     6    1  test1    x    30
2     8    2  test2    x    50
3     8    2  test3    w    45
4     8    1  test2    y    35
5     4    2  test3    y    20
6     4    1  test3    y    20
7     2    1  test3    z    22
8     2    2  test3    y    32

The created anonymized data is below(Guarantee 2-anonymity).

  col1 col2         col3 col4  count
0  2-4    2        test3    y      2
1  2-4    1        test3    y      1
2  2-4    1        test3    z      1
3  6-8    1  test1,test2    x      2
4  6-8    1  test1,test2    y      1
5    8    2  test3,test2    w      1
6    8    2  test3,test2    x      1

anonypy's People

Contributors

Stargazers

Watchers

anonypy's Issues

Example in README returns error

I'm hoping to learn how to use this package, but when I run the example in the README, I get the following error:

---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
Input In [2], in <cell line: 1>()
----> 1 main()

Input In [1], in main()
     26 sensitive_column = "col4"
     28 p = anonypy.Preserver(df, feature_columns, sensitive_column)
---> 29 rows = p.anonymize_k_anonymity(k=2)
     31 dfn = pd.DataFrame(rows)
     32 print(dfn)

File ~/miniconda3/envs/ppla-dev2/lib/python3.8/site-packages/anonypy/anonypy.py:18, in Preserver.anonymize_k_anonymity(self, k)
     17 def anonymize_k_anonymity(self, k):
---> 18     return self.__anonymize(k)

File ~/miniconda3/envs/ppla-dev2/lib/python3.8/site-packages/anonypy/anonypy.py:10, in Preserver.__anonymize(self, k, l, p)
      8 def __anonymize(self, k, l=0, p=0.0):
      9     partitions = self.modrian.partition(k, l, p)
---> 10     return anonymize(
     11         self.modrian.df,
     12         partitions,
     13         self.modrian.feature_columns,
     14         self.modrian.sensitive_column,
     15     )

File ~/miniconda3/envs/ppla-dev2/lib/python3.8/site-packages/anonypy/anonypy.py:79, in anonymize(df, partitions, feature_columns, sensitive_column, max_partitions)
     75 grouped_columns = df.loc[partition].agg(aggregations, squeeze=False)
     76 sensitive_counts = (
     77     df.loc[partition].groupby(sensitive_column).agg({sensitive_column: "count"})
     78 )
---> 79 values = grouped_columns.iloc[0].to_dict()
     80 for sensitive_value, count in sensitive_counts[sensitive_column].items():
     81     if count == 0:

AttributeError: 'list' object has no attribute 'to_dict'

Unable to use package

When trying to use the anonpy package, I get the error:

AttributeError: 'list' object has no attribute 'to_dict'on the anonimize function: anonymize(df, partitions, feature_columns, sensitive_column, max_partitions)