Spring-Spark-K-Anonymity

This repository contains a Spring-boot application with services exposed to generate and anonymize data sets in a distributed manner.

Description

Implements K-Anonymization in a distributed manner using:

• Apache Spark for partitioning and merging of data set.
• Executor Service for asynchronous anonymization of partitions.
• ARX library for applying K-Anonymization on partitioned data set.

Prerequisites

1. Install Java as described in How do I install Java?
2. Install Maven as described in Installing Apache Maven
3. Install Apache Spark in Spark Overview

Project Structure

├───java
│   └───com
│       └───spark
│           ├───arx 
│           ├───car_basic
│           ├───controllers
│           ├───data_generator
│           ├───file_utils
│           └───spark_config
└───resources
    ├───input-data
    └───output-data

Build and Run

• Build the project by installing all dependencies required using the below command:

mvn clean install

• Run the spring-boot application using the below command:

mvn spring-boot:run

The application will run on port 8080 by default.

Data Set Generation

1. For generating data set, run the below Standalone Java application:

javac DataSetGenerator.java
java DataSetGenerator.java

2. To increase the number of records in the data set, the below-mentioned strategies can be used:
   • Increasing the number of cars.
   • Capturing latitude and longitude for shorter distance intervals.

Anonymization

1. Below endpoints are exposed as part of the Spring application.

Method	Request	Description
POST	/api/health	Reports if the application has started.
POST	/api/anonymize	Anonymizes the created data set.

2. The number of partitions can be altered as shown below.

sparkConfigurator.loadDataSource(carSchema, 20);

3. The K-value can be specified as shown below.

config.addPrivacyModel(new KAnonymity(10));