|
Title:
|
APPLICATION OF NEURAL NETWORK TO PREDICT ADVERSE SITUATIONS IN TROUBLE TICKETING REPORTS |
|
Author(s):
|
Julia Gómez , Yaiza Temprado , Margarita Gallardo , Carolina García , Francisco Javier Molinero |
|
ISBN:
|
978-972-8924-87-4 |
|
Editors:
|
António Palma dos Reis |
|
Year:
|
2009 |
|
Edition:
|
Single |
|
Keywords:
|
Trouble ticketing, severity, escalation, artificial neural networks |
|
Type:
|
Short Paper |
|
First Page:
|
204 |
|
Last Page:
|
208 |
|
Language:
|
English |
|
Cover:
|
|
|
Full Contents:
|
click to dowload 
|
|
Paper Abstract:
|
The volume and variety of information stored in databases have dramatically grown in the last decade. Specifically, in
large companies' infrastructures, thousands of trouble tickets can be generated every year. In these companies with a huge
network infrastructure, it is critical to optimize the response that is given whenever a failure is encountered. It is essential
to take advantage of that information in order to find and extract patterns which allow to predict these behaviours and be
ahead of the situation, as well as to be able to describe tendencies and regularities.
This paper describes the project, developed by Telefónica I+D, about the application of an artificial neural network in
order to predict the severity changes that trouble tickets suffer during their life cycle. For this purpose, a study of the
problem and the neural network that better adapts to these conditions are presented. |
|
|
|
|
|
|
