User-Guided Exploration of Large Datasets using M-Trees

Theoretical (Analytical):

Practical (Implementation):

Literature Work:


Exploring large amounts of data using data visualizations is a commonly performed tasks by analysts and scientists. However, guiding scientists or analystis to interesting parts of the data is still a hard and time-consuming problem. To help experts identifying promising parts of the data and to speed up the exploration process and M-Tree based data structure is utilizied.

Problem Statement

  • Exploration of large datasets can be unintuitive and time-consuming.
  • Visualization and interaction designs rely on suitable background data structures.


  • Adapt the M-tree
    • Remove the database index constraints
    • Split criterion should be determined by node size but rather by the data
  • Create meaningful visualizations to help experts explore the adapted M-tree and allow user interaction.
  • Design a prototypical framework.    


  • Basic knowledge about information visualization
  • Good programming skills in Java


  • Scope: Bachelor/Master
  • 6 Month Project, 3 Month Thesis (Bachelor) / 6 Month Thesis (Master)
  • Start: immediately



  • Ciaccia, Paolo; Patella, Marco; Zezula, Pavel (1997). "M-tree An Efficient Access Method for Similarity Search in Metric Spaces"