Visualizing discovered knowledge in databases mining

Other Title(s)

تخيل المعارف المكتشفة في تنقيب قواعد البيانات

Dissertant

Abu Khudayr, Yasir Ishaq Dawud

Thesis advisor

al-Ani, Muzhir Shaban

Comitee Members

al-Hamami, Ala Husayn
Aqil, Misbah Jumah

University

Amman Arab University

Faculty

Collage of Computer Sciences and Informatics

Department

Department of Computer Science

University Country

Jordan

Degree

Master

Degree Date

2011

English Abstract

Extracting knowledge from educational data delivers advanced business intelligence and web discovery solutions.

The education sector in Jordan faces a significant challenge to growth while Jordan’s business and government leaders ascertain the country’s differentiated competitive advantage relative to other regional and global players.

In order to focus marketing and promotion efforts—as well as to develop a strategy to expand and diversify Jordan’s higher education offering, there is a need to better understand the availability and capacity of the labor supply side that will match the demand expectations of companies seeking to locate higher-and higher education operations within Jordan.

One of the biggest challenges that higher education faces today is predicting the paths of students.

Institutions would like to know, for example, which universities; its students are the most employed.

Are some students more likely to work than others? In addition to this challenge, statistics are extremely important for development plans, forecasting trends, and for business process outsourcing.

We propose an approach for classifying the most suitable major based on historical data, based on decision tree method, and visualizing knowledge discovered from educational database, by adopting ID3 decision tree algorithm and incorporating Dashboard concept for Jordanian higher education institution.

Experiments were run for educational databases associated with different institutions, beginning with Ministry of Higher Education (MOHE), National Information Technology Center (NITC), public and private universities, and Social Security Corporation in Jordan.

In this thesis we are trying to find out classes of workers, using ID3 decision tree algorithm a classification algorithm, that produce a tree that is used to classify future samples of data, in order to help companies, to extract classes of students who are most probably to work.

In order to achieve our goals, we process various sources of databases (a total of 250000 records), and partitioned it into training and testing data, and finally applying ID3 decision tree algorithm.

Additionally, a group of statistical reports are implemented, in order to help decision makers to make decisions.

The detected or inferred information should increase our knowledge about each major, university, and faculty.

The resulted decision tree will be used to apply the testing data on.

A set of structural models from different universities, and majors have been in our database.

Different structural models have been proposed, with statistical reports.

The average accuracy of the models is close to (79%) using a lift chart.

Main Subjects

Information Technology and Computer Science

Topics

• Management
• Databases
• Data mining

No. of Pages

112

Table of Contents

• APA
• MLA
• AMA

Language

English

Data Type

Arab Theses

Record ID

BIM-525952