README.md

Data-Mining-Big-Data-2017-Project

Data Mining - Big Data Project 2017 - Uncertain Clustering.
U-DBSCAN is a density based algorithm that can cluster uncertain data using as distance measure a probabilistic distance between the random distributions of the points.
Usage:

1. Download the U-DBSCAN.jar file
2. If it is the first time that you run the algorithm and you want to use an existing dataset, create a new folder in your documents named "bdmpFiles" and inside it create a folder named "input".
   If, in your first execution, you don't want to use an existing dataset simply go to 4) point.
3. Put "dataset.txt" and "datasetInfo.txt" in the input folder
4. Execute it with "java -jar U-DBSCAN.jar"
5. Output will be saved into "bdmpFiles/output" folder