Two-party Jointly Decision Tree Algorithm

An extended version of decision tree for privacy data mining

Download

The whole project can be download from: https://github.com/monaen/PrivacyDecisionTree

Requirements

java JDK >= 1.8.0

Features

• Extend the traditional ID3 algorithm to two-party jointly build situation.
• Can fit the privacy situation better.

Drawbacks

• Can only implement on the dataset which only contains two target classes.
• Cannot deal with the case of missing value of attribute.
• Data structure is space-consuming.

Contributors

• Nan Meng

How to use

Download from github

Run git clone https://github.com/monaen/PrivacyDecisionTree.git

Run the project at local

cd PrivacyDecisionTree/src
javac PrivacyID3.java
java PrivacyID3

Algorithm illustration

• Privacy Decision Tree Algorithm