JOURNAL OF COMPUTING, VOLUME 3, ISSUE 4, APRIL 2011, ISSN 2151-9617

HTTPS://SITES.GOOGLE.COM/SITE/JOURNALOFCOMPUTING/
WWW.JOURNALOFCOMPUTING.ORG 61

Techniques developed in Artificial Intelligence

from the standpoint of their application in
software Engineering
A.Sharmila Dhana Joy, Dr. R.Dhanapal
Abstract—Software development process is very complex process is primarily a human activity. Programming, in software
development, requires the use of different types of knowledge: about the problem domain and the programming domain. It
also requires many different steps in combining these types of knowledge into one final solution. This paper intends to
study the techniques developed in artificial intelligence (AI) from the standpoint of their application in software engineering.
In particular, it focuses on techniques developed (or that are being developed) in artificial intelligence that can be deployed
in solving problems associated with software engineering processes. This paper highlights a comparative study between
the software development and expert system development. This paper also highlights absence of risk management strate-
gies or risk management phases in AI based systems.

Index Terms – Knowledge intensive activity, Programmer’s apprentice, automated programming, genetic code.

——————————  ——————————

1 INTRODUCTION

A rtificial Intelligence is concerned with the study

and creation of computer systems that exhibit some
form of intelligence and attempts to apply such know-
ledge to the design of computer based systems that can
understand a natural language or understanding of natu-
ral intelligence. Software Engineering is a knowledge-
intensive activity, requiring extensive knowledge of the
application domain and of the target software itself.
Many Software products costs can be attributed to the
ineffectiveness of current techniques for managing this
knowledge, and Artificial Intelligence techniques can help
alleviate this situation.
The goal of this research paper is to give a comparative
study of how expert programmers analyze, synthesize,
modify, explain, verify and document programs, and to
apply that theory towards automating the programming
process. Recognizing that the long-term goal of totally
automatic programming is very far off, we are presently
concentrating on applying our research towards develop-
ing an intelligent computer assistant for programmers,
————————————————
 A.Sharmila Dhana Joy is with the Department of Computer Applica-
tion,Research Scholar,Dravidian university, Andhra Pradesh,India
 Dr. R.Dhanapal is with the Department of Computer Application,
Easwari Engg. College, Affiliated to Anna University, Chennai, India
called the Programmer’s apprentice.
One of my key observations is that expert programmers
rely heavily on a large body of standard implementation
methods and program forms. A central part of the re-
search has therefore been to identify and codify these
standard forms. For this purpose many expert system
programming languages have been developed in which
these standard forms can be written down in a canonical
and abstract way, and used by an automatic program-
ming systems. Basically Conventional programming is a
sequential, three step process. Design, Code, Debug.
Knowledge engineering, which is the process of building
an expert system, also involves assessment, knowledge
acquisition, design, testing, documentation and mainten-
ances. However, there are some key differences between
the two programming paradigms. Conventional pro-
gramming focuses on solution, while ES programming
focuses on problem.
2 SOFTWARE ENGINEERING AND ARTIFICAL
INTELLIGENCE
Software development
Software development problems includes conceptual spe-
cifying, designing, testing the conceptual construct and

© 2011 Journal of Computing Press, NY, USA, ISSN 2151-9617

http://sites.google.com/site/journalofcomputing/
JOURNAL OF COMPUTING, VOLUME 3, ISSUE 4, APRIL 2011, ISSN 2151-9617
HTTPS://SITES.GOOGLE.COM/SITE/JOURNALOFCOMPUTING/
WWW.JOURNALOFCOMPUTING.ORG 62

representation problems that comprising representing  System evaluation

software and testing the reliability of a representation.
The traditional view of software development process
begins at the requirements specification and ends at test-
ing the software. At each of these stages, different kinds
of knowledge (design  knowledge at design stage and
programming and domain knowledge at the coding
stage) are required. At each of the two stages: design and
coding, exist a cycle: error recognition and error correc-
tion. Experience show that errors can occur at any stage
of software development. Errors due to coding may oc-
cur because of faulty design. Such errors are usually ex-
pensive to correct.
A basic problem of software engineering is the long delay
between the requirements specification and the delivery
of a product. This long development cycle causes re-
quirements to change before product arrival.
Planning phase involves feasibility assessment, resource
allocation, task phasing and scheduling requirements
analysis. Knowledge acquisition is the most important
stage in the development of ES. During this stage the
knowledge engineer works with the domain expert to
acquire, organize and analyze the domain knowledge for
the ES. The goal of knowledge analysis is to analyze and
structure the knowledge gained during the knowledge
acquisition phase. After knowledge analysis is done, we
enter the knowledge design phase. At the end of design
we have Knowledge definition, detailed of a development
tool. Consider whether it supports your planned strategy,
internal fact structure, Mock interface. Coding this phase
occupies the least time in the Expert System Development
Life Cycle. It involves coding, preparing test cases, com-
menting code, developing user’s manual and installation
guide.

In addition, there is the problem of phase independence Knowledge based techniques in AI can be used to modify
of requirements, design and codes. Phase independence this traditional approach the AI technique that handles
means that any decision made at one level becomes fixed this problem is automated programming which results in
for the next level. Thus, the coding team is forced to re- reusable code. Thus, when a change is made in the de-
code whenever there is change in design. sign, that part of the design that does not change remains
unaffected. Thus, automated tools for system redesign
Programing
and reconfiguration resulting from a change in the re-
Design
Knowledge Knowledge quirements will serve a useful purpose. This technique
requires constraint propagation technique. With the help
of automated programming approach AI based systems
Requirements
Specification Design
Design
Testing
Coding Code
Testing
are free from risk management strategies.

Planning Knowldge Knowldge Code Knowldge System

Acquisition Design Verificaion Evaluaion
and
Analysis

Error
Detection Error Work Encoding of
Detection Knowledge Design Knowledge Actual
Baseline Baseline using a testing Product
development evaluation
Fig.1: Traditional Software Development Process Process by user
tool

Expert system development

Expert system use knowledge rather than data to control Fig 2: Expert System development
the solution process. Knowledge engineers build systems
by eliciting knowledge from experts, coding, that know-
ledge, and ultimately constructing a system using a varie-
ty of building tools. 3 TECHNIQUES AND TOOLS OF AUTOMATED
The main phases the expert system development PROGRAMMING
processes are:-
 Planning
Because of the evolutionary nature of software products,
 Knowledge acquisition and analysis
by the time coding is completed, requirements would
 Knowledge design
have changed (because of the long processes and stages of
 Code development required in software engineering): a situa-
 Knowledge verification

© 2011 Journal of Computing Press, NY, USA, ISSN 2151-9617

http://sites.google.com/site/journalofcomputing/
JOURNAL OF COMPUTING, VOLUME 3, ISSUE 4, APRIL 2011, ISSN 2151-9617
HTTPS://SITES.GOOGLE.COM/SITE/JOURNALOFCOMPUTING/
WWW.JOURNALOFCOMPUTING.ORG 63

tion that results in delay between requirement specifica- out the product development phase.
tion and product delivery. There is therefore a need for
design by experimentation, the feasibility of which lies in The given Figure displays the steps of the risk manage-
automated programming. Some of the techniques and ment process. Formally, articulated, risk management
tools that have been successfully demonstrated in auto- process consists of three steps:
mated programming environments include Language
Feature: this technique adopts the concept of late binding
(i.e. making data structures very flexible). In late binding,
data structures are not finalized into particular imple-
mentation structures. Thus, quick prototypes are created
which result in efficient codes that can be easily changed. Risk
Management
Another important language feature is the packaging of
data and procedures together in an object, thus giving rise
to object-oriented programming: a notion that has been
found useful in environments where codes, data struc-
Risk Risk Risk
tures and concepts are constantly changing. Lisp provides Identification Analysis Mitigation
these facilities.
• Meta Programming: this concept is developed in natu-
ral language processing (a sub field of AI). It uses auto-
mated parser generators and interpreters to generate ex- Risk
probability
Risk
Avoidance
ecutable lisp codes. Its use lies in the modelling of transi-
tion sequences, user interfaces and data transformations.
Risk
• Program Browsers: these look at different portions of a Risk Impact Monitoring
code that are still being developed or analyzed, possibly
to make changes, thus obviating the need for an ordinary Contingen-
cy Plan-
text editor. The browser understands the structures and Risk Factor
ning
declarations of the program and can focus on the portion
of the program that is of interest.
• Automated Data Structuring: this means going from a Fig 3 : Risk Management Process
high-level specification of data structures to a particular
implementation structure.
AI based systems are free form risk management strate-
When systematic changes need to be made throughout a
gies because of automated programming techniques mak-
code, it is more efficient and controllable to do it through
ing data structures flexible. Automatic programming is
another program (i.e., program update manager) than
the generation of programs by computer, usually based
through a manual txt editor. For instance, a change in
on specifications that are higher – level and easier for
program X may be required whenever h is being updated
humans to specify than ordinary programming languag-
by b-1 under the condition that b is less than C. Assume
es.
that a program W makes a systematic change in all such
places. If another program makes a change in W, then any
program changed by W also must be updated. Thus, pro-
gram update managers propagate changes. Because of
Requirement Automatic Executable
this ability, program update managers are useful when
Specification Programing System Program
prototypes need to be developed quickly

Fig 4: Automatic Programming system (APS)

4 RISK MANAGEMENT
The Risk Management process is a method of identifying
risks in advance and establishing methods of avoiding
those risks and / or reducing the impact of those risks
should they occur.
The Process of risk management begins during the analy-
sis phase of software development life cycle. However,
the actual process of managing risks continues through-
The goal is to make the specification smaller, easier to
write, easier to understand (closer to application con-
cepts), Less error-prone better than programming lan-
guages.
Genetic Code
Genetic programming is a technique which enables com-
puters to solve problems without being explicitly pro-
grammed. It works by using genetic algorithms to auto-

© 2011 Journal of Computing Press, NY, USA, ISSN 2151-9617

http://sites.google.com/site/journalofcomputing/
JOURNAL OF COMPUTING, VOLUME 3, ISSUE 4, APRIL 2011, ISSN 2151-9617
HTTPS://SITES.GOOGLE.COM/SITE/JOURNALOFCOMPUTING/
WWW.JOURNALOFCOMPUTING.ORG
matically generate computer programs.
Evaluate each of the attempted solution
Keep the best solutions
Produce next generation from these solution (using mu-
tation and crossover)
Quit when you have a satisfactory solution (or you run
out of time)
Genetic algorithm contains a population of trail solutions
to a problem. In genetic programming the individuals in
the population are computer programs.
5 CONCLUSION
Risk management strategies utilize lot of developer time
and in software development phases there is a link be-
tween all the phases by introducing a isolation phase
among the phases we can reduce the time in development
by revisiting each phase after changes in requirements. By
using AI based systems with the help of automated tool or
automated programming tool we can eliminate risk as-
sessment phase saving our time in software development.
Because of Artificial Intelligence Techniques implemented
in software engineering. We can reduce the development
time in software development. Coding phase in software
development process can be changed into Genetic Code.
REFERENCES
1] Ian Sommerville, Software Engineering (6th Edn.) (Addision
Wesley Publishers, Newyork, New York, USA) 2000
[2] Roger S.Pressman, Software Engineering: A Beginner’s Guide
(Mc Graw Hill Higher Education Publishers, New York, New York,
USA) 1998
[3] Seth Hock, Computers and Computing (Houghton Mifflin
College Publishers, Boston, Massachusetts, USA) 1989
[4] Shari Lawrence P fleeger, Software Engineering: theory and
Practice (Prentice Hall Publishers, Upper Saddle River, New Jersey,
USA) 1998.
[5] Artificial Intelligence (3rd Edition) Prentice Publisher,
Henry Petrick Winston
[6] T.H. Tse, T.Y. Chen, R.L. Glass, An assessment of systems and
software engineering scholars and institutions (2000–2004), Journal
of Systems and Software 79 (6) (2006) 816–819.
[7] C. Wohlin, An analysis of the most cited articles in software
engineering journals –1999, Information and Software Technology
Vol. 47 (No. 15) (2005) 957–964.
[8] M. Zitt, S. Ramanan-Rahary, E. Bassecoulard, Relativity of cita-
tion performance and excellence measures: from cross-field to cross-
scale e ects of field normalisation, Scientometrics Vol. 63 (No. 2)
© 2011 Journal of Computing Press, NY, USA, ISSN 2151-9617
http://sites.google.com/site/journalofcomputing/
64
(2005) 373-401
[9] Fenton NE, Neil M A critique of software defect prediction
models IEEE Transactions on Software Engineering 25 (5): 675–689
Sep–Oct 1999
[10] Dr.R.Dhanapal, “An Intelligent Information Retrieval Agent”.
Elsevier International Journal on Knowledge-Based Systems, The
Netherlands, Vol. 21 No. 6, pp. 466 – 470, 2008.
AUTHOR BIBILIOGRAPHY
A. Sharmila Dhana joy is with the Department of Computer Applica-
tion, Vel Sri Ranga Sanku college, Affiliated to Madras University ,
Chennai – 600 005, Tamil Nadu, India, a research scholar pursuing
PhD in Computer Science from Dravidian University .
Dr.R.Dhanapal obtained his PhD in Computer Science from Bhara-
thidasan University, India. He is currently Professor, PG and Re-
search Department of Computer Applications, Easwari Engineering
College, Chennai. Affiliated to Anna University of Technology Chen-
nai, Tamil Nadu India. He has 25 years of teaching, research and
administrative experience. Besides being Professor, he is also a
prolific writer, having authored twenty one books on various topics in
Computer Science. His books have been prescribed as text books in
Barathidasan University and autonomous colleges affiliated to
Bharathidasan University, Tiruchiraplli. He has served as Chairman
of Board of Studies in Computer Science of Bharathidasan Univer-
sity, member of Board of studies in Computer Science of several
universities and autonomous colleges. Member of standing commit-
tee of Artificial Intelligence and Expert Systems of IASTED, Canada
and Senior Member of International Association of Computer Sci-
ence and Information Technology (IACSIT), Singapore. He has Vis-
ited USA, Japan, Malaysia, and Singapore for presenting papers in
the International conferences and to demonstrate the software de-
veloped by him. He is the recipient of the prestigious ‘Life-time
Achievement’ and ‘Excellence’ Awards of Govt. of India. He is serv-
ing as Principal Investigator of UGC sponsored innovative, major
and minor research projects about 1.7 crore. He is the recognized
supervisor for research programmes in computer Science leading to
PhD and MS by research in several universities including Anna Uni-
versity of Technology Chennai, Bharathiar University Coimbatore,
Manonmaniam Sundaranar University Tirunelveli, Mother Teresa
University Kodaikanal and many Deemed Universities. He has got
52 papers on his credit in international and national journals. He has
been serving as Editor In chief for the International Journal of re-
search and reviews in Artificial Intelligence (IJRRAI) United Kingdom
and serving as reviewer and member of editorial in accredited peer
reviewed national amd international journals including Elsevier Jour-
nals.