Universität Bonn: Autonomous
    Intelligent Systems Institute for Computer Science VI: Autonomous Intelligent Systems  

ILP, the Blind, and the Elephant:
Euclidean Embedding of Co-Proven Queries

Abstract

Relational data is complex. This complexity makes one of the basic steps of ILP difficult: understanding the data and results. If the user cannot easily understand it, he draws incomplete conclusions. The situation is very much as in the parable of the blind men and the elephant that appears in many cultures. In this tale the blind work independently and with quite different pieces of information, thereby drawing very different conclusions about the nature of the beast. In contrast, visual representations make it easy to shift from one perspective to another while exploring and analyzing data. In this thesis, we describe the first method for embedding the contents of a relational database and the rules governing the data into a single, common Euclidean space based on their co-proven statistics. The embedding is designed to place objects which are often co-proven close to each other, while unrelated objects are placed far from one another. We analyze the properties of the embeddings and demonstrate our method on real-world datasets, showing that ILP results can be intuitively represented and indeed be captured at a glance.

Documentation

  • The paper, presented on ILP 2009
    @inproceedings{elephant09,
    title = { {ILP, the Blind, and the Elephant: Euclidean Embedding of Co-Proven Queries} },
    author = {Schulz, H and Kersting, K and Karwath, A},
    booktitle = {ILP},
    publisher = {Springer},
    series = {Lecture Notes in Computer Science},
    note = {to be published},
    year = {2009}
    }
        
  • The thesis, which explains the method in more detail
  • The talk, which, however, doesn't contain much text
  • The spotlight, a 2-slide presentation

Sample Images