الفهرس 
AppendixOnly 14 pages are availabe for public view
 |  | from | 158 |  |  |
| | | from | 158 | | |
Abstract
Insight problem solving using heuristics is a promising approach of artificial
intelligence that is based on Gestalt research on cognition. The first known
story that link insight problem with heuristics is about Archimedes. ‘Insight’
can mean the profound understanding of a field of study, but we will restrict it
to the sudden discovery of how a problem works. Insight is usually symbolized
by a glowing light bulb to stress the sudden illumination of the insight. This
thesis tries to answer the following questions: Can a simple problem become an
insight problem? And how? Can one type of insight problems be transformed
to another? Intuitively, can heuristics help in solving insight problems? What
are the proofs both theoretically and practically ? What are the main sources of
heuristics? Can we reach a heuristic generator?
First, the thesis investigates an example for mathematical insight problem
using symmetry, another example for verbal insight problem by trying to solve
twins problem using logic programming languages, and an example for spatial
insight problem by trying to solve optical illusion.
The thesis also introduces an opportunity for applying insight in braincomputer
interaction (BCI). The Pied Piper story inspired us to devise new
heuristics for interfacing human motor system (HMS) using brain waves. By
combining head helmet and LumbarMotionMonitor (LMM) and distinguishing
between knowledgeable and not knowledgeable signals, it is possible to
produce solutions for handicaps.
Finally, the thesis proposes many nature-inspired metaheuristic optimization
algorithm in different fields such as social networks, data mining, robotics, and
neuroscience. Let us present a couple of them:
There are two patterns of random walks: Brownian movement is
associated with abundant preys and Lévy flight is associated with
sparser or unpredictably distributed preys. With apparent potential
capability, Lévy flight is applied to optimal search. Hoopoe Heuristic
(HH) generates random moves using Lévy-Flight. After a while, the
Hoopoe Heuristic changes its behavior by generating random moves using Ground-Probing. This may be interpreted by the level of
experience a hoopoe has after a time in investigating the landscape. It
outperforms other algorithms, with 100% optimality and less running
time on standard test functions.
There is a need for new metaphors from immunology to flourish the
application areas of Artificial Immune Systems. A metaheuristic called
Obesity Heuristic derived from advances in obesity treatment is
proposed. The main forces of the algorithm are the generation omega-6
and omega-3 fatty acids. The algorithm works with Just-In-Time
philosophy; by starting only when desired. A case study of data cleaning
is provided. With experiments conducted on standard tables, results
show that Obesity Heuristic outperforms other algorithms, with 100%
recall. This is a great improvement over Genetic algorithms.