Javascript must be enabled for the correct page display

A Primal Support Vector Machine for Handwritten Character Recognition using a Bag of Visual Words

Wanningen, A. (2016) A Primal Support Vector Machine for Handwritten Character Recognition using a Bag of Visual Words. Bachelor's Thesis, Artificial Intelligence.

[img]
Preview
Text
AI_BA_2016_AnneWanningen.pdf - Published Version

Download (386kB) | Preview
[img] Text
Toestemming.pdf - Other
Restricted to Backend only

Download (77kB)

Abstract

Using a bag of visual words for unsupervised feature learning, a system of handwritten character recognition is developed using a support vector machine (SVM) for which the update rules are derived directly from the primary objective function. The system is tested on the MNIST dataset and outperforms a traditional SVM implementation in this particular set-up, not only in terms of accuracy but most notably in terms of computation time. Whether this approach should be labelled as part of the SVM family of algorithms remains a point of discussion.

Item Type: Thesis (Bachelor's Thesis)
Degree programme: Artificial Intelligence
Thesis type: Bachelor's Thesis
Language: English
Date Deposited: 15 Feb 2018 08:25
Last Modified: 15 Feb 2018 08:25
URI: http://fse.studenttheses.ub.rug.nl/id/eprint/14697

Actions (login required)

View Item View Item