Machine Learning Using Topology Signatures For Associative Memory

Abstract

This thesis presents a technique to produce signatures from topologies generated by the Growing Neural Gas algorithm. The generated signatures have the following characteristics: The signature's memory footprint is smaller than the "real object" and it represents a point in the n x m multidimensional space. Signatures can be compared based on Euclidean distance and distances between signatures provide measurements of differences between models. Signatures can be associated with a concept and then be used as a learning step for a classification algorithm. The signatures are normalized and vectorized to be used in a multidimensional space clustering. Although the technique is generic in essence, it was tested by classifying alphabet and numerical handwritten characters and 2D figures obtaining a good accuracy and precision. It can be used for many other purposes related to shapes and abstract typologies classification and associative memory. Future work could incorporate other classifiers.

Author Keywords: Associative memory, Character recognition, Machine learning, Neural gas, Topological signatures, Unsupervised learning

    Item Description
    Type
    Contributors
    Creator (cre): Florez, Elkin Dario
    Thesis advisor (ths): McConnell, Sabine
    Thesis advisor (ths): Hurley, Richard
    Degree granting institution (dgg): Trent University
    Date Issued
    2015
    Date (Unspecified)
    2015
    Place Published
    Peterborough, ON
    Language
    Extent
    150 pages
    Rights
    Copyright is held by the author, with all rights reserved, unless otherwise noted.
    Subject (Topical)
    Local Identifier
    TC-OPET-10236
    Publisher
    Trent University
    Degree