(IJACSA) International Journal of Advanced Computer Science and Applications,

Vol. 12, No. 5, 2021

Automating and Optimizing Software Testing using

Artificial Intelligence Techniques
Minimol Anil Job
Faculty of Computer Studies, Arab Open University
Kingdom of Bahrain

Abstract—The final product of software development process technologies, various business domains are accepting and
is a software system and testing is one of the important stages in using AI based software. AI systems are developed based on
this process. The success of this process can be determined by machine learning models and techniques. AI is used to
how well it accomplishes its goal. Due to the advancement of promote automation and reduce the amount of routine
technology, various software testing tools have been introduced activities to create testing phases by applying logic, problem
in the software engineering discipline. The use of software is solving, and ML. The purpose of this paper is to present the
increasing day-by-day and complexity of software functions are application of AI techniques in automation of software testing
challenging and there is need to release the software within the
and the impacts. The paper is closed with a conclusion and a
short quality evaluation period, there is a high demand in
viewpoint of future work to enhance the practical aspects of
adopting automation in software testing. Emergence of automatic
software testing tools and techniques helps in quality
the AI automation tool.
enhancement and reducing time and cost in the software II. LITERATURE REVIEW
development activity. Artificial Intelligence (AI) techniques are
widely applied in different areas of Software engineering (SE). Testing is an activity takes place throughout the software
Application of AI techniques can help in achieving good development life cycle. The major aim of the software testing
performance in software Testing and increase the productivity of is find errors and to ensure that the developed system satisfies
the software development firms. This paper briefly presents the the requirements of customers. Software testing a way of
state of the art in the field of software testing by applying AI evaluating a system by discovering the differences between
techniques in software testing. the identified requirements at the requirement engineering
activity and the archived results. There are various techniques
Keywords—Software testing; artificial intelligence; testing used for testing a software to ensure quality. Testing can be
automation; software engineering; software quality done as manual or automated using specific tools [20] [7]. In
I. INTRODUCTION manual testing, the software tester follows a test plan and
complete a set of test cases manually. Manual testing is time
Software engineering is the creation and application of consuming and cost effective. The alternative of the manual
sound engineering principles to produce cost-effective testing is the automated execution in which a software
software which is both reliable and functional on real program runs a pre-defined test cases and shows and saves the
machines. A well-managed development process is needed to results. Automatic testing executes the test-case automation,
produce a high-quality software product. Software test-case generation and result verification. [2]. Automation
development is a human endeavor which involves various testing tools are used in automatic software testing and the
activities. The activities are; analysis, design, implementation reliability as well as performance are more than manual
and testing and each of them contributes to the creation of the testing [10] [19]. Software testing using AI techniques has
final product. These activities continue in the development been adopted by software development companies all over the
process, and thus producing a working version of the system world. The application of AI is wide-ranging branch of
can be time-consuming. Software testing is one of the computer science that deals with building smart machines
principal activities in software development for verifying and which are capable of performing smart tasks with assistance of
validating a software system. Testing assists software human intelligence [9].
developers in ensuring that the developed software fulfills its
intended function, as well as determining whether or not the AI has become an emerging technology which can be
identified problems have been solved. Since the software applied in ensuring quality assurance (QA) in organizations
development life-cycle is a complex process with a crucial [26] [15]. AI developments require an approach to validation
need to deliver a new product within the allocated time, the and verification of software. AI has becoming an important
software testing process should be efficient and effective. factor in quality assurance and testing in future. [14]. AI is a
branch of science and engineering concerned with the
In the software industry, automation plays a critical role in computational analysis of intelligent behavior and the creation
increasing testing performance. There are various automation of intelligent machines. The term AI refers to a collection of
tools available to support the testing activity. Newer tools, techniques, and algorithms [3] [11] [21]. Application of
technologies like Artificial Intelligence (AI) and Machine AI techniques are creating impacts in various domains like
Learning (ML) are constantly being used to speed up the health, manufacturing etc. Emergence of big data and its
software development process. With the advancement of AI management requires strong computing tools and AI helps in

594 | P a g e
www.ijacsa.thesai.org
 (IJACSA) International Journal of Advanced Computer Science and Applications,
Vol. 12, No. 5, 2021

improving the efficiency of these activities. AI techniques are A. Categories of Testing

applied in different areas of software engineering such as Software developers need to be aware and to be able to use
requirement engineering, design and testing [24]. AI in a range of other testing methods, concepts and practices. The
software testing aims to make testing smarter and more four distinct categories of testing are summarized in Fig. 1:
efficient. AI and machine learning apply reasoning and
problem solving to automate and improve testing. AI in
software testing helps reduce time-consuming manual testing,
so teams can focus on more complex tasks, like creating
innovative new features. Faulty software is mission critical as
software becomes an increasingly important component in a
wide variety of systems. Since automated test generation
techniques are free of the cognitive biases that have been
found in human testers, they can be used for software testing
[4].
Automating the execution process of the software testing
cycle is the most common approach in the automation field.
Automated software testing is important and adding more
toolkits to make testing phases fully automated by generating
test scripts is also equally important. Test cases can be
generated to complete the automated testing [22]. The test
execution speed will be increased by this tools and the testing
Fig. 1. Categories of Testing.
process can be applied repeatedly. In manual testing method,
developing test scripts is a time consuming process but if the Table I presents the details of the four categories of testing
test cases are ready the human tester can complete the testing shown in the above figure.
process quickly [27] [25]. In manual testing, code visibility
does not affect test code coverage and fault detection rate12]. TABLE I. TYPES OF TESTING
It is possible to facilitate a process for eliciting testing
requirements and creating test-suites [1]. Various algorithms REQUIREMENTS-BASED TESTING
are used in AI tools. Use of Bayesian probabilistic reasoning  Checks that a system meets the customer's requirements using previously
to model software reliability is an example probabilistic AI gathered or formulated testable requirements.
technique. This is used in the discipline of Software  Acceptance testing is the final stage to check that the user requirements
Engineering [18] [5] [21]. AI techniques can be applied in have been satisfied.
 The customer formally accepts the software by ensuring the correct
supporting automation and decreasing the amount tedious functioning of the acceptance tests
tasks in the software development and testing phase [17].
USABILITY TESTING
III. SOFTWARE TESTING  User interface (UI) is an integral part in every software system.
Software testing is one of the crucial stage in the software  Various users other than the developers are using the system and the
usability testing is essential.
development life cycle and it is an important method to assess
 Due to the problems of user interface, the systems may fail.
the developed software to ensure quality. Before releasing the
final product, it is important to ensure that the requirements
 Usability testing comprises systematically trying out the user interface
with intended users
identified during the analysis stage have been accomplished as
well as the product is defect free. In the context of software DEVELOPMENTAL TESTING
development, bug is a name given for the defect which means  Refers to the testing carried out by the entire software development team.
the developed system is not producing accurate results as per  The developmental testing is done in three different levels which are unit
the requirements. Testing plays an essential part in the testing, integration testing and system testing.
software development process. Various techniques can be  Integration testing involves checking that all the tested units are interfaces
used during the software development process to facilitate together correctly.
testing. Prototyping is one of the approaches in which an REGRESSION TESTING
experimental software artefact will be developed and that will  Any form of testing done during the development or maintenance of a
be discarded after evaluation. Early software artefacts are built system to ensure that fixing one bug does not result in the introduction of
on rather than discarded in the Iterative approach [23]. new ones.
Another approach is using frameworks such as the dynamic  are needed at unit, integration and system levels depending on the
systems development method (DSDM) in which best practice particular development method adopted
processes for iterative and incremental development are  Regression tests are critical during the developmental testing and in
system maintenance.
documented [13].

595 | P a g e
www.ijacsa.thesai.org
 (IJACSA) International Journal of Advanced Computer Science and Applications,
Vol. 12, No. 5, 2021

TABLE II. BLACK-BOX TESTING TECHNIQUES

B. Software Testing Techniques
Black-Box Testing Techniques
Strategies for creating test cases is an important part of
validation and verification. The two major strategies are Equivalence partitioning
black-box testing and white-box testing. The main objective of  This technique divides the input domain of the software unit into groups
black box testing is to ensure each aspect of the customer’s or partitions and generates test cases thus helps in reducing the number of
requirements is handled correctly by an implementation. The test cases.
test cases are designed by looking at the specification of the Boundary value analysis
system. The specification includes the details of the systems  Errors at the boundaries of the input domain are tested in this technique.
intended functions.  The boundary values are usually taken the minimum and maximum values
in the boundaries and values just inside and outside of the boundaries.
Grey-Box Testing is another method in which the Orthogonal Array Testing
application is tested with complete knowledge of the overall
 It's a strategy for problems with a limited input domain that are relatively
aspects of the system and limited knowledge of the internal
small and cannot be tested thoroughly
functioning of the system. [12]. The major objective of white
box testing is to check that the details of an implementation of All pair Testing
the system are correct. The test cases are designed by looking  In this method, test cases are created using all possible combinations of
at the detail of the implementation of the system to be tested to each pair of input parameters.
check that the system performs its intended functions correctly Cause and effect Graph
[8].
 In this approach, a graph is generated and a cause-and-effect relationship
While comparing the possibility of automation of black- is formed. The cause is the input condition that leads to internal change in
box and while-box testing techniques, automating white-box the system and the effect is the output condition.
testing is easier than black-box testing. The reason is that box
in black-box testing, programmer and the tests are dependent TABLE III. WHITE-BOX TESTING TECHNIQUES
and this will affect the automation. Since grey-box testing White-Box Testing Techniques
provides the features of both black-box and white-box testing
techniques, it’s features and techniques are not considered in Control Flow Testing
detail in this paper.  This technique uses a testing strategy in which control flow of the
program is involved in the test coverage. Branch coverage, statement
Fig. 2 depicts the five strategies used in black-box testing coverage and condition coverage are the three methods used in test
and four strategies used in white-box testing. coverage.
Data Flow Testing
The different black-box testing techniques are summarized
 Data movement within the program is focused in this testing strategy and
in Table II [17, 22]. the test paths are determined from the positions of variable definition in
The different approaches of White box testing are listed in the program. This technique helps to detect errors such as undefined
variables etc.
Table III [17, 22].
Loop Testing
 Different types of loops such as simple loops, nested loops are used in
most of the programs. Validity of the loops in the programs are tested
using this technique.
Basic Path Testing
 Basis-path testing is based on the cyclomatic-complexity metric. The
number of independent paths in a method body is computed using this
metric.

IV. TEST AUTOMATION

The need to make sure that users are really receiving the
value promised by the software, many powerful testing tools
have emerged, which make it easy to do black box testing in
an automated and repeatable manner. The automation is
important because the testing process is repetitive and it is
recommended to test all possible scenarios. Automating tests
will increase the test coverage and improve efficiency. When
the testing process is automated, the tests can be executing
repeatedly, test different input values and various conditions.
The testing resources and time will be reduced. To automate
functional, system and acceptance tests, numerous tools are
available. Some of them are Selenium, Watir and JMeter.
Selenium and Watir are used to test local files using the file://
protocol supported by web browsers, however, JMeter needed
Fig. 2. Black-Box and White-Box Testing Techniques. the target files to be hosted on a web server.

596 | P a g e
www.ijacsa.thesai.org
 (IJACSA) International Journal of Advanced Computer Science and Applications,
Vol. 12, No. 5, 2021

V. APPLICATION OF AI IN SOFTWARE TESTING

Due to the increasing complexity and features, modern
applications require various features in order to achieve the
functional and non-functional requirements of the
applications. Requirements are the information about what a
system will be and do that needs to be known before
development starts. Once the requirements are identified and Fig. 4. Functions in a Manual Testing Environment.
agreed by the developer and the customer, they will be
implemented in the software system. Converting the
requirements into an application using appropriate tools and
techniques is the role of the system developer. The application
should be developed without errors and during the coding
stage, the developer needs to write down and execute test
cases. A software tester is a person who is in charge of putting
an application through its paces and ensuring that it performs
as intended. Software testers play a crucial role same as the
developer in the software engineering activity as they are part
of the quality assurance of the software. The software tester Fig. 5. Functions in an AI Enables Testing Environment.
sends a report to the development team detailing the bugs
found and the sequence of events that contributed to the In a manual testing environment, the tester runs the test
mistake. As shown in Fig. 3, the Developer‘s responsibility of cases according to the test plans, finds the bugs and reports
writing code and writing and executing test cases. The testers them to the developer. In the AI based testing environment,
will follow the test plans and complete the testing tasks. the tester runs test cases and finds the bugs. The AI tool do the
Software testing is an important process which ensures the diagnosis and send notification to the developer to fix the
developed system’s satisfaction from the customers. Testing bugs. The AI tool plans further tests automatically and the
helps to protect the system against any failure which may lead testers will continue running the test cases and identify bugs.
potential impact in organizations’ performance during the The functional and non-functional requirements of the
operational period. In manual testing, the software is tests are product need to be tested by ensuring the full test coverage. In
executed by the software tester to discover bugs in the system. manual testing this will take time and resources and
Even though exploratory testing is possible in manual testing, application of AI techniques will reduce the time and also will
it is a time-consuming process by involving human resources help to identify poorly designed requirements. Requirements
and test cases are executed by a human tester and software. will be thoroughly verified using AI and it will be transferred
Automated testing is considerably faster than manual testing to the design stage, the next stage in the software development
and faster than a manual approach and uses automation tools life-cycle. Based on the experience of the software testers, the
to execute test cases. As testing is moving more and more test selection and quality will vary and the tests are created
towards automation, AI techniques are applied in software based on the requirements, use cases and user stories.
testing. AI techniques supports the automation of the testing Application of AI helps to automatically generate test cases
processes in an efficient way. AI algorithms are supporting the and identify the tests for verification and validation of the
testing environment to be more productive. The AI algorithm system. Ai also help to understand the test coverage as well.
encourages the process and helps software testers to find the Missing or inappropriate test can be identified based on the
maximum number of bugs within less period of time. The models. Ain addition to this, automatic generation of the test
system can be route to market as reliable and accurate. Fig. 4 cases will also reduce any eluded bugs and defects also.
and Fig. 5 show the functions of a tester and a developer in
manual and AI enables testing environments. VI. APPLICATION OF AI IN TEST AUTOMATION
Business organizations in various sectors are adopting AI
techniques makes their operations efficient. In software
development process, automation testing is widely used to
improve the test efficiency. There are various ways AI
techniques can support the software testing process [6] [16].
The most benefit takes from AI is the black box testing.
Applications of AI techniques in software testing is listed.
Fig. 6 depicts the test automation in pyramid shape
according to the test types. UI testing is placed in the highest
level because and this is the most expensive testing. Test
automation can be done in the lower levels both in API and
Fig. 3. Tasks: Software Developer Vs Software Tester. Unit layers to achieve best test coverage and reduce time and
cost.

597 | P a g e
www.ijacsa.thesai.org
 (IJACSA) International Journal of Advanced Computer Science and Applications,
Vol. 12, No. 5, 2021

• aid in the prediction of incorrect test cases that result in

total test failures, as well as include recommendations
for resolving those issues.
• used in simulating Behavioral patterns.
o AI techniques helps to simulate the behaviors of
people in using the system by geography,
devices and demographics as inputs to build test
suites.
• AI algorithms are efficient in prediction and
automation of test suites.
Fig. 6. Test Automation Levels. • used in script Automation.
o Automating a test script is not required in AI
User Interface (U) testing: Usability tests are conducted to
identify any design inconsistencies or usability issues in the because it is executed from the AI algorithm
user interface. To build UI tests, AI techniques use image automatically.
recognition techniques to navigate through the application and
• Helping in mining defects and based on this test suite
visually validate UI objects and components. The test
can be defined. This speeds up the process of making
automation tools decode the Document Object Model (DOM)
informed decisions about test coverage and test suite
and related code to determine object properties in AI-based UI
optimization.
testing. The UI tests are done after the design and
development of the software system. In the pyramid, UI tests • Enhances number of tests and their scopes.
are in the top level of the pyramid and the automation is
challenging and they are the hardest to automate. • AI techniques are applied in Automation Test
Maintenance, Test Data Generation, early feedback in
Application Program Interface (API) Testing: In API testing etc.
testing, validation of Application Programming Interfaces
(APIs) is done by testing various software quality factors of VII. AI BASED TEST AUTOMATION TOOLS
the programming interfaces such as functionality,
performance, reliability and security. Nonfunctional A. AI Bases Automation Tools
requirements like look-and-feel are not relevant in API testing Table IV summarizes 5 AI based automation tools uses in
and focuses on the business logic instead. Without any software industry recently.
reference to the user interface, API automation can be used to
validate the business logic. Automation of the API allows the TABLE IV. AI BASED AUTOMATION TESTING TOOLS
software development team to start the testing activity early Testing Tool Application
which helps to identify and fix errors and reducing the efforts
in the AI testing. This AI tool is used to automate functional testing by
using artificial intelligence and machine learning. This
Testim
Creating and updating unit tests: Unit tests are used to automated testing tool speeds up the authoring, execution,
ensure that small pieces of code, such as functions or object and maintenance of automated tests.
methods, are correctly executed before they are combined A cloud based automated testing technology used for
with other units to form a larger functional feature. Developers Functionize
functional, performance and load testing. This automated
spend a significant amount of time writing and maintaining testing tool speeds up the test creation, diagnosis, and
unit tests, which is much less enjoyable than writing maintenance.
application code AI-based products for automated unit test This AI enabled tool used for automating functional,
development can be beneficial and they can be generated Appvance performance and security testing. The AI integrated tool
promptly. Automation strategy starts with unit tests. Unit tests helps codeless test creation.
are written by the developers using the same application This AI enabled tool used for User Interface (UI) testing.
development programming language. Early detection and It provides an end-to-end software testing platform
Applitools
correction of errors in the unit testing will prevents regression powered by Visual AI and used in Quality Assurance and
test automation.
defects.
This AI-powered test automation platform supports
In addition to the above mentioned testing automation regression testing and also for monitoring of web
Testcraft
strategies AI can be applied in the following areas of the applications. The AI technology helps in eliminating
software testing also. maintenance time and cost.
This is an open Source tool, uses Ruby Libraries to
• AI is used as an adaptive method for detecting element- automate tests. Watir is used for testing websites and uses
level changes that will improve the testing suites' Watir Selenium for browser automation. In addition to this
robustness. Watir also supports in writing stable and maintainable test
scripts.

598 | P a g e
www.ijacsa.thesai.org
 (IJACSA) International Journal of Advanced Computer Science and Applications,
Vol. 12, No. 5, 2021

Functional tests are part of integration tests and designed any web application can be automated irrespective of the
to evaluate specific tasks. For example, user registration in an browser it is running. There are in-built libraries in Watir to
online system, the registration function should validate the support various activities such as page performance. Fig. 8
data entry from the user. Other examples are catalogue search depicts the test script for opening Chrome browser.
and online payment in an e-commerce application etc. A wide
range of test values are required to perform these tests. Fig. 9 demonstrates another sample of creation and
Performance testing is used for measuring the ability of execution of test case using Watir. A simple test has been used
servers to respond to user demand in a networking to test the accuracy of the code. The test script runs
environment. A load test is a method of evaluating a server's successfully.
performance by applying a given load to the application in Fig. 10 shows an example of a test script for the user
order to evaluate the response time for individual functions. interface testing. The sample HTML and JavaScript code is
The aim of security testing is to ensure that the application has shown in Fig. 11 for the page.
authentication and authorization controls in place and is not
vulnerable to attacks. User interfaces are usually tested for The presentation of the web page is done using HTML and
Navigation errors, presentation errors and control usage a JavaScript function is used for the static feature of the page
problems. Regression testing can reveal defects that recur as a code using HTML and Fig. 11 shows the code.
result of software changes that were not planned. Fig. 12 shows the debugging screen of another sample test
B. Pactical Application: Using Watir script with another browser, Firefox using Watir automation
tool.
Watir (Web Application Testing in Ruby) has been
selected as a software experimenting too to support the
technical aspects of the research. l. The Watir supports almost
all of the web browsers. The websites and user interfaces can
be tested using Watir. The test framework by using Watir
follows the following order as shown in Fig. 7.
After installing the required drivers, test scripts for
opening browsers have been written in the RubyMine IDE.
The test scripts are written in Ruby language. Watir is selected
for the practical application because first of all, it is very easy
to use open source tool. This tool is developed using Ruby and Fig. 7. The Order of Test Framework in Watir.

Fig. 8. Test Script for Chrome Browser.

Fig. 9. Sample Test Script.

599 | P a g e
www.ijacsa.thesai.org
 (IJACSA) International Journal of Advanced Computer Science and Applications,
Vol. 12, No. 5, 2021

Fig. 10. Test Script for user Interface Testing.

Fig. 11. HTML and JavaScript Code for the Sample UI.

Fig. 12. Debugging Screen.

600 | P a g e
www.ijacsa.thesai.org
 (IJACSA) International Journal of Advanced Computer Science and Applications,
Vol. 12, No. 5, 2021

After running the test cases using it can be concluded that activity to produce a defect free software application. This
Watir allows easy to test file download for the user interface paper presented the role of artificial Intelligence tools in
of a web application. Popup test alerts are provided at times as software testing.
well. Another feature in Watir is, Page Performance can be
measured using the performance object. Navigation, timing The paper also looked into various types of testing
and memory performance can be measured by getting these techniques for validation and verification of the software
details when the application is connected to the browser. Page system. Selection of an appropriate AI based test automation
object feature available in Watir helps reusability of the code tools is important based on the type of testing and this paper
basically in the form of classes. This feature of Watir helps the discusses the features of popular AI enables testing tools.
automation of the application without code redundancy. Finally, the paper is presented the advantages of AI testing
Without opening the browser, the details are obtained in the tools applications in software testing. An extension of this
command line and thus supports the execution of the user research paper will be presented by evaluating AI based test
interface test cases at the command line prompt. automation tools by taking into consideration of more
technical details. Enhanced practical aspect will be covered by
VIII. ADVANTAGES OF USING AI IN SOFTWARE TESTING real implementation of the test automation of white-box
testing using a mobile application.
Fig. 13 lists the advantages of using AI tools in automating
software testing. When the testing of a software system is REFERENCES
done manually by even by highly skilled and experienced [1] Agrafiotis, I., Creese, S., Goldsmith, M.: Developing a Strategy for
software testers there is a possibility of making mistakes and Automated Privacy Testing Suites. In: Camenisch J., Crispo B., Fischer-
Hübner S., Leenes R., Russello G. (eds.) Privacy and Identity
experience tiredness. Artificial Intelligent tools are constantly Management for Life. Privacy and Identity 2011. IFIP Advances in
performing tasks effectively even repetitive tasks. Automated Information and Communication Technology, vol. 375, pp. 32–44.
testing tools cab ne used by both software developers and Springer, Berlin, Heidelberg (2012). doi:10.1007/978-3-642-31668-5_3
testers. If the modifications ad editing of the program source 2.
code are cheeked thoroughly software tests will run [2] Andreas Leitner, Ilinca Ciupa, Bertrand Meyer, Mark Howard
automatically. For any unsuccessful tests, the developers will “Reconciling Manual and Automated Testing: the AutoTest
Experience”. ETH Zurich CH-8092 Zürich, AXA Rosenberg Investment
be notified and thus saving the time of the developers. AI Management LLC Orinda, California 94563.
enables automatic testing can help in improving the overall [3] A. M. Turing. “Computing machinery and intelligence”. In: Parsing the
test coverage and ensure quality of the developed software. Turing Test. Springer, 2009, pp. 23–65. [2] J. McCarthy. “Artificial
The performance of the system can be tested against the intelligence, logic and formalizing common sense”. In: Philosophical
expected requirements in an efficient manner. AI enables logic and artificial intelligence. Springer, 1989, pp. 161–190.
testing can perform image and pattern recognition. This [4] Berndt, D.J., Fisher, J., Johnson, L., Pinglikar, J., and Watkins, A.,
feature will help in detecting visual errors and make sure all “Breeding Software Test Cases with Genetic Algorithms,” In
Proceedings of the Thirty-Sixth Hawaii International Conference on
visual elements are working properly. Software testing is a System Sciences (HICSS-36), Hawaii, January 2003.
repeating process and source code may change time to time, [5] B. Littlewood and J. L. Verrall, “A Bayesian reliability growth model
manual testing will be a time consuming task. AI tools can for computer software,” Applied Statistics, vol. 22, no. 3, pp. 332–346,
perform the testing and detect errors with lesser time and AI 1973.
tools are not expected to make errors and will help to generate [6] E. Horvitz, J. Breese, D. Heckerman, D. Hovel, and K. Rommelse, “The
more accurate results. Test automation can be done efficiently Lumiere project: Bayesian user modeling for inferring the goals and
AI enables testing tools. needs of software users,” in Proceedings of the Fourteenth Conference
on Uncertainty in Artificial Intelligence. San Mateo: Morgan Kaufmann,
Jul. 1998, pp. 256–265.
[7] Er. Rajender Bathla, Er. Shallu Bathla. “Innovative Approaches of
Automated tools in Software testing & current technology as compared
to Manual testing ”. Global Journal of Enterprise Information System,
Vol-1(1) , 2009.
[8] Glenford J. Myers, Corey Sandler, Tom Badgett , The Art of Software
Testing, 3rd Edition, 2015.
[9] Harman, M. (2012, June). The role of artificial intelligence in software
engineering. In 2012 First International Workshop on Realizing AI
Synergies in Software Engineering (RAISE) (pp. 1- 6). IEEE.
Fig. 13. AI Enabled Testing: Advantages. [10] Jagdish Singh, Monika Sharma.” A Comprehensive Review of Web-
based Automation Testing Tools”. International Journal of Innovative
IX. CONCLUSION AND FUTURE WORK Research in Computer and Communication Engineering, Vol. 3(10),
October 2015.
Producing a quality software to the clients within the [11] J. McCarthy. Artificial intelligence: a paper symposium: Professor Sir
specified time by incorporating all the requirements is a James Lighthill, FRS. Artificial Intelligence: A General Survey. In:
crucial task. By using annual testing approach this will not be Science Research Council, 1973. 1974.
easy. There are various automation tools available to perform [12] Khan, M. E., & Khan, F. (2012). A comparative study of white box,
software testing. Artificial Intelligence techniques have black box and grey box testing techniques. Int. J. Adv. Comput. Sci.
Appl, 3(6).
significant impact in various stages of software development
[13] Lee Copeland, A Practitioner’s Guide to Software Test Design, 2003.
activity including software testing. Application of AI in test
automation is an appropriate solution for software testing [14] M. Buenen and A. Walgude, “World quality report 2018–19,” Paris,
France, Tech. Rep., 2018.

601 | P a g e
www.ijacsa.thesai.org
 (IJACSA) International Journal of Advanced Computer Science and Applications,
Vol. 12, No. 5, 2021

[15] Mariani, L., Hao, D., Subramanyan, R., Zhu, H.: The central role of test [21] S. Russell et al. Artificial Intelligence: A Modern Approach. 3rd ed.
automation in software quality assurance. Software Quality Journal Prentice Hall, 2010.
25(3), 797–802 (2017). doi:10.1007/s11219-017-9383-5. [22] S. H. Trivedi, "Software Testing Techniques," International Journal of
[16] Meziane, F. and Vadera, S., (2010). Artificial Intelligence in Software Advanced Research in Computer Science and Software Engineering,
Engineering Current Developments and Future Prospects, In "Artificial vol. 2, no. 10, pp. 433439, 2012.
Intelligence Applications for Improved Software Engineering [23] Stuart Feldman, Quality Assurance: Much More than Testing, ACM
Development: New Prospects", IGI Global. Digital Library, Queue – Quality Assurance, February 2005, Vol 3, Issue
[17] Narayan, Vaibhav, The Role of AI in Software Engineering and Testing 1.
(June 22, 2018). International Journal of Technical Research and [24] Tadapaneni, N. R. (2017). Different Types of Cloud Service Models.
Applications, 2018, Available at SSRN: https://ssrn.com/abstract= Available at SSRN 3614630.
3633525.
[25] T. Kosa, M. Mernikb and T. Kosarb, "Test automation of a measurement
[18] N. E. Fenton, M. Neil, W. Marsh, P. Hearty, L. Radlinski, and P. system using a domain-specific modelling language," Journal of
Krause, “On the effectiveness of early life cycle defect prediction with Systems and Software, vol. 111, p. 74–88, 2016.
Bayesian Nets,” Empirical Software Engineering, vol. 13, no. 5, pp.
499–537, 2008. [26] Van De Ven, T. et al., 2018. World Quality Report 2018-19 - Artificial
Intelligence - World Quality Report 2018-19 Findings:. [Online]
[19] Prof. (Dr.) V. N. MAURYA , Er. RAJENDER KUMAR . “Analytical Available at: https://www.sogeti.com/globalassets/global/wqr-
Study on Manual vs. Automated Testing Using with Simplistic Cost 201819/wqr-201819_secured.pdf.
Model ”. International Journal of Electronics and Electrical Engineering,
vol .2 (1), 2012. [27] V. Sangave and V. Nandedkar, "A Review on Automating Test
Automation," International Journal of Advance Research in Computer
[20] R. M. Sharma.” Quantitative Analysis of Automation and Manual Science and Management Studies, vol. 2, no. 12, 2014.
Testing.” International Journal of Engineering and Innovative
Technology (IJEIT), Vol. 4(1), 2014.

602 | P a g e
www.ijacsa.thesai.org