Grammar based hyper-heuristic

Dissertant

Abd al-Muttalib, Muhammad Adnan Fuad

Thesis advisor

Bittaz, Muhammad
Khatri, Sunil Kumar

University

Philadelphia University

Faculty

Faculty of Information Technology

Department

Department of Computer Science

University Country

Jordan

Degree

Master

Degree Date

2019

English Abstract

The development of proper algorithmic solution for a given class of problems requires a deep understanding of some optimization algorithms and this process is time consuming.

In this study, we investigate the hyper-heuristic methodology which is a high-level search methodology that operates on search space of heuristic/meta-heuristic algorithms.

Hyper-Heuristic aims at finding the most suitable algorithmic solution for a given class of problems.

Hyper-heuristic is classified into two major classes: selective and generative hyper-heuristic.

Our focus is on generative hyper-heuristic, especially on generative hyper-heuristic that operates on the meta-heuristic components of Particle Swarm Optimization (PSO) and Real-Coded Genetic Algorithm (RCGA).

The study uses a modified Tree-based Grammar-guided Genetic Programming (TG3P), in order to generate adaptive hybrid PSO and RCGA solvers for continuous global optimization problems.

We compared our results with two prominent PSO algorithms, and the results show that our proposed hyper-heuristic has very competitive efficiency.

Main Subjects

Information Technology and Computer Science

Topics

• Algorithms
• Genetic programming

No. of Pages

66

Table of Contents

• APA
• MLA
• AMA

Language

English

Data Type

Arab Theses

Record ID

BIM-955861