Please use this identifier to cite or link to this item:

http://hdl.handle.net/10609/52612
Title: Open source, web-based machine-learning assisted classification system
Author: Consarnau Pallarés, Mireia Roser
Director: Viejo Galicia, Luis Alexandre
Others: Universitat Oberta de Catalunya
Keywords: machine learning
classification
open source
Issue Date: 23-Jun-2016
Publisher: Universitat Oberta de Catalunya
Abstract: The aim of this article is to provide a design overview of the web based machine learning assisted multi-user classification system. The design is based on open source standards both for multi-user environment written in PHP using the Laravel framework and a Python based machine learning toolkit, Scikit-Learn. The advantage of the proposed system is that it does not require the domain specific knowledge or programming skills. Machine learning classification tasks are done on the background automatically. The paper commences with a review of literature on applications of text mining the most common type of data available, discusses the main system components and outlines the process flow with examples. System effectiveness is also evaluated using a dataset comprised of music lyrics divided into two classes of 245 songs each, using a support vector machine (SVM) classifier, function words feature set, three fold cross-validation and 10:90 train test split.
Language: English
URI: http://hdl.handle.net/10609/52612
Appears in Collections:Bachelor thesis, research projects, etc.

Share:
Export:
Files in This Item:
File Description SizeFormat 
mconsarnauTFM0716memòria.pdfMemòria del TFM219.38 kBAdobe PDFView/Open

Items in repository are protected by copyright, with all rights reserved, unless otherwise indicated.