To analyse incompleted data based on "distance" and visualize clusters of related data.
- Generate random incomplete data.
- Make this data available via and http server.
- Split the data into groups consisting of entries that shares missing fields.
- display the id's of each entrie and the group it relates to.