EEG data analysis by using map-reduce technique

Other Title(s)

تحليل بيانات مخطط كهربائية الدماغ (EEG)‎ باستخدام تقنية Map-Reduce

Dissertant

Fulayyih, Ali Adil

Thesis advisor

al-Hamami, Ala Husayn

Comitee Members

Utair, Muhammad Abd Allah
al-Sarhan, Hasan Muayyadi

University

Amman Arab University

Faculty

Collage of Computer Sciences and Informatics

Department

Department of Computer Science

University Country

Jordan

Degree

Master

Degree Date

2016

English Abstract

Database is defined as a set of data, which is organized and distributed in a manner that make it easy to user to access these data in more efficient way.

With rapid growth in data capturing and with concurrently in developing of power processing, data transmission, storage capability enable users to integrate various Databases in centralized data management for maximizing the vision to access, analysis, retrieval which called Data Warehouse and then it leads to emerge Data Mining which is used for revealing and extracting hidden precious value form large Databases (Data Warehouse).However, the arising and rapid growing in the size of data sets make it more complex that it is about 85% of these data is unstructured and semi structured.

To deal with the problem of big-size and more complex data, a new branch of IT has been emerged, which is called Big-Data.

In the era of Big-Data, the traditional methods of data analytics may not be suitable and efficient to manage and process the large amount and complex of data.

This thesis aimed to study the use of Map-Reduce processing technique to handle Big-Data distributed on the Cloud-Computing, and to develop an efficient way to handle these Big-Data.

The work was evaluated using Hadoop server, and applied on Electroencephalogram (EEG) Big-Data as a case study.

Using Hadoop in Cloud-Computing as an environment for this kind of applications is so efficient for-at least- four reasons: the highly fault tolerance it has, the automated data distributed it performs, the balancing of the computation load across different nodes it performs, parallel computation property it has and as close as possible the computation location from data position property it has that reflects in network overhead of transferring.

The proposed approach was to transform EEG text file to list then can deal whit it due to the fact that it has index, which formerly it was impossible to extract valuable information for EEG researchers.

The experiment result showed clear enhancement on managing and processing the EEG big data with average of 50% reduction on response time.

The obtained results provide EEG researchers and specialist with an easy and fast method of handling the EEG data.

Main Subjects

Information Technology and Computer Science

No. of Pages

152

Table of Contents

• APA
• MLA
• AMA

Language

English

Data Type

Arab Theses

Record ID

BIM-722669