Unsupervised and Supervised Learning with the Random Forest Algorithm for Traffic Scenario Clustering and Classification

Published in IEEE IV 2019

Setup

1. DOWNLOAD the Matlab files from Github (*.m and *.p files)

2. DOWNLOAD the dataset as a zipped file from google drive (337 MB):
   https://drive.google.com/file/d/1voRvjNlhVArO-r8k1ZJBc6v_Mf5tjSKh/view?usp=sharing

3. COPY all Matlab files and the dataset into one folder

4. OPEN Main_RUN.m in Matlab

5. RUN directly, or adapt the variables according to your needs.

   • Further information is given in the Main_RUN.m file

Data

The zipped file contains two *.mat files for the plotting function and two *.mat files with the dataset used for the clustering and classification process.

Feature Description

A description of the 47 features is given in the FeatureDescription.m file.

Compatibility

Tested on Matlab R2017b, R2018a, R2019a