Reference:

Jaakko Peltonen. Visualization by linear projections as information retrieval. In José Príncipe and Risto Miikkulainen, editors, Advances in Self-Organizing Maps (proceedings of WSOM 2009), pages 237–245, Berlin Heidelberg, 2009. Springer.

Abstract:

We apply a recent formalization of visualization as information retrieval to linear projections. We introduce a method that optimizes a linear projection for an information retrieval task: retrieving neighbors of input samples based on their low-dimensional visualization coordinates only. The simple linear projection makes the method easy to interpret, while the visualization task is made well-defined by the novel information retrieval criterion. The method has a further advantage: it projects input features, but the input neighborhoods it preserves can be given separately from the input features, e.g. by external data of sample similarities. Thus the visualization can reveal the relationship between data features and complicated data similarities. We further extend the method to kernel-based projections.

Suggested BibTeX entry:

@inproceedings{Peltonen09wsom,
    address = {Berlin Heidelberg},
    author = {Jaakko Peltonen},
    booktitle = {Advances in Self-Organizing Maps (proceedings of WSOM 2009)},
    editor = {Jos{\'e} Pr{\'i}ncipe and Risto Miikkulainen},
    pages = {237-245},
    publisher = {Springer},
    title = {Visualization by Linear Projections as Information Retrieval},
    year = {2009},
}