UUM ETD | Universiti Utara Malaysian Electronic Theses and Dissertation
FAQs | Feedback | Search Tips | Sitemap

Uncovering Hidden Information Within R&D Department's Ticket Using Data Mining Clustering Approach

Azmi, Abu Bakar (2011) Uncovering Hidden Information Within R&D Department's Ticket Using Data Mining Clustering Approach. Masters thesis, Universiti Utara Malaysia.

[img] PDF
Azmi_Abu_Bakar.pdf
Restricted to Registered users only

Download (779kB)
[img]
Preview
PDF
1.Azmi_Abu_Bakar.pdf

Download (163kB) | Preview

Abstract

MP Company's R&D department in Penang received daily submission tickets (this represent issues raised by its staffs) from it staff requesting for support. Number of annual tickets submission increases from year 2007 until 2009. Increase of issues means that increase of support activities in order to resolve these extra issues. Directly this will increase the cost of operation. This project will undergo analysis which prescribes in one of data mining technique called Clustering analysis. Hidden information and major root cause of the increase issues is expected to be unveiled. Result of this analysis can be used to generate framework or solution to improve the situation and stabilized the number of tickets submission. In this study the data extracted is clustered using two different types of data mining techniques i.e. K-Means and Kohonen Network. Later the clustered produced is compared and evaluated using Multinomial Logistic Regression and Neural Network: MLP. The result produced then reveals the biggest root caused of issue or problems that eventually triggered the ticket being submitted. This knowledge will be used by MP Company to further produce the framework or solution model for implementation.

Item Type: Thesis (Masters)
Subjects: Q Science > QA Mathematics > QA76 Computer software
Divisions: College of Arts and Sciences (CAS)
Depositing User: Mr Husni Ismail
Date Deposited: 21 Nov 2011 07:37
Last Modified: 27 Apr 2016 02:17
URI: http://etd.uum.edu.my/id/eprint/2501

Actions (login required)

View Item View Item