A hybrid approach using fuzzy logic and MapReduce to achieve meaningful using of big data

Dissertant

al-Mukahhal, Ikhlas Hasan

Thesis advisor

al-Fayyumi, Muhammad
al-Ziyadat, Wail

University

Isra University

Faculty

Faculty of Information Technology

Department

Department Software Engineering

University Country

Jordan

Degree

Master

Degree Date

2018

English Abstract

Big data is facing many challenges in different aspects, which appears through characteristics such as: volume, velocity and variety.

Processing and analyzing big data are challenging issues to acquire quality information to support accurate values for decision making.

Quality data taxonomy point to three basic elements which are meaningfulness, predicatively, and decision making, where all of these elements encourage investigating previous work that focused on the same challenges of big data.

Consequently, the Hybrid Approach using Fuzzy Logic and MapReduce produces a new version of MapReduce consisting of four layers with data collection in the first layer, while the second layer preprocesses and treats semi-structured data clean up.

Map function to acquire relationships, applying fuzzy controller and classification to generate rules among occur data in the third layer, whereas the last layer includes reduce and classification to achieve a predicative and meaningful outcome.

Experiment A Hybrid Approach uses Diabetes patient data and symptoms in R (Readr, Dplyr, Tidyr, PreProcess, HadoopStreaming, HiveR and FuzzyR ) packages and Weka; Evaluation confirmed significant efficiency of the proposed approach through precision, recall and F-measure.

Main Subjects

Information Technology and Computer Science

Topics

• Fuzzy logic
• Data

No. of Pages

64

Table of Contents

• APA
• MLA
• AMA

Language

English

Data Type

Arab Theses

Record ID

BIM-838475