|
Title:
|
PARAMETER-SETTING FREE HARMONY SEARCH OPTIMIZATION OF RESTRICTED BOLTZMANN MACHINES AND ITS APPLICATIONS TO SPAM DETECTION |
|
Author(s):
|
Luis A. da Silva, Kelton A. P. da Costa, Patricia B. Ribeiro, Gustavo Rosa, João Paulo Papa |
|
ISBN:
|
978-989-8533-45-6 |
|
Editors:
|
Hans Weghorn |
|
Year:
|
2015 |
|
Edition:
|
Single |
|
Keywords:
|
Spam Detection, Machine Learning, Restricted Boltzmann Machines, Optimum-Path Forest. |
|
Type:
|
Full Paper |
|
First Page:
|
143 |
|
Last Page:
|
150 |
|
Language:
|
English |
|
Cover:
|
|
|
Full Contents:
|
click to dowload 
|
|
Paper Abstract:
|
Spam detection has been one of the foremost machine learning-oriented applications in the context of computer networks. In this paper, we propose to learn intrinsic properties of e-mail messages by means of Restricted Boltzmann Machines (RBM) in order to identity whether such messages contain relevant (ham) or non-relevant (spam) content. The main contribution of this work is to employ Harmony Search-based optimization techniques to fine-tune RBM parameters, as well as to evaluate their robustness in the context spam detection. The unsupervised learned features are then used to feed the Optimum-Path Forest classifier, being the original features extracted from e-mail content compared against the new ones. The results have shown RBM are suitable to learn features from e-mail content, since they obtained promising results in one out of the two public datasets employed in this work. |
|
|
|
|
|
|
