Published on

Published in: Technology, Business
  • Be the first to comment

  • Be the first to like this

No Downloads
Total views
On SlideShare
From Embeds
Number of Embeds
Embeds 0
No embeds

No notes for slide


  1. 1. ISSN: 2277 – 9043 International Journal of Advanced Research in Computer Engineering & Technology Volume 1, Issue 2, April 2012 SOFTWARE USABILITY EVALUATION METHOD Piyush Jain , Sanjay Kumar Dubey , Prof.(Dr.)Ajay RanaAbstract - Software usability is a important factor in We have also tried to include the usabilityevaluating the quality of any software. All the software engineering phase in the SDLC (Softwarequality model has consider the usability as one of the Development Life Cycle) to improve themost important criteria of judging the performance of performance of software. Although we alwaysany software. Different methods have been employed tocalculate usability, but in this paper we are providing a consider it in our SRS document but it is needed to bevery simple technique in this paper. This paper also represented as a distinguished phase because of theprovides a method by which we can compare the user specification. The software are designed for theperformance of different software also. Software user , so it is required to always keep a track in everyusability engineering is an area which deals with the phase of SDLC. In this paper we have discussedusability evaluation of any software. The usability about a very simple way to estimate the usability ofworkers try to shape up the user requirement within any software [2]. The evaluation method is beensoftware. implemented by using fuzzy logic [2] which deals with uncertainty. Sometimes we have some complexThis paper deals with the evaluation of different problem where the final result is uncertain orusability external metrics which helps the performance unpredictable. In that case we use soft techniques forof software and then fuzzifying those metrics Fuzzy dealing with this problem. Soft techniques deals withlogic is probabilistic reasoning system that deals with human intelligent system like study of human brainapproximation rather that fixed or exact calculation [4].As the user requirement changes from time to time it is for faster computation , human genetic system forrequired to design which adapt itself with the changes. carrying the characters of parents generation byIt can be possible with the help of fuzzy logic which generation and taking decision in a state ofrepresent the facts with a truth value which varies from uncertainty. A soft technique also has the capability0-1. Fuzzy logic has the capability to represent the to adapt itself with any change in the system bycomplex system in a simple rule based system. employing different learning technique.Index Term - Fuzzy logic, Linguistic Variable, In this paper we will introduce and work with one ofFuzzification , Software Usability, Estimation. soft technique ie is dealing with uncertainty known as fuzzy logic. To explain how fuzzy logic works just I. INTRODUCTION take a example to measure the room temperature without using thermometer. In that case we willThis paper has tried to evaluate the usability and estimate it by using the inference rule system like “Ifquality of software by using fuzzy logic system. we Sweat , then temperature in room is High”. In thisSoftware usability is an factor of software quality example we have to design two fuzzy set “Sweat”model which deals with the user requirement. The and “High” known as linguistic variable. Now weuser requirement is specified by using a USD (user will associate a truth value with the entire element inspecification document) used by the software this fuzzy set which will vary from 0-1. Based on thatengineers to specify the requirement of user from the we will evaluate the temperature of the room .expected designed software. .Piyush Jain Computer Science Amity UniversityNoida,India,+91-9873818831 The paper is organized in the following way. SectionSanjay Kumar Dubey Computer Science Amity University I gives an introduction to the paper and section IINoida, India, introduces the concept of Fuzzy logic. Section IIIProf Ajay Rana Computer Science Amity University deals with the evaluation methods used till now andNoida, India 28 All Rights Reserved © 2012 IJARCET
  2. 2. ISSN: 2277 – 9043 International Journal of Advanced Research in Computer Engineering & Technology Volume 1, Issue 2, April 2012the last section ie section IV introduces the different in fuzzy set , fuzzy inference rule for outputmetrics which are under consideration based on formulation and then defuzzifier for outputwhich the evaluation will be performed . calculation. A simple fuzzy system can be designed as described in the figure 1. II. FUZZY LOGIC Step1. Consider the problem and design the fuzzy setFuzzy logic was first coined by Zadeh A. Lofti in based on the linguistic terms.1960 in his seminal paper [6][7]. This employs a Step2. Define some fuzzy rules .natural way to deal with the problem where the Step3. Calculate the fuzzy weighted metrics based onsource of information is uncertain. In fuzzy logic all the concept.the problem are solved by designing the fuzzy set Step4. Convert the fuzzy system into a single scalarwhich are defined for the linguistic terms. The fuzzy crisp value. This is known as defuzzification.system uses fuzzifier to convert the real time problem INPUT DATA TO THE DEFINE THE CALCULATE THE FUZZY SYSTEM INFERENCE RULE WEIGHTED DEFUZZIFICAT MATRICS ION •Fuzzification • Rule based system • Calculation of Scalar output Fig.1 Fuzzy System III. DIFFERENT USABILITY iv. Perceptive based Inspection : It has EVALUATION METHOD designers who represent multiple other projects The usability engineering is an area which deals with inspect an interface to see whether it does things inthe user requirement specification. Generally for the same way as their own designs.evaluating software usability we uses the three  Usability Testing: In this method thefollowing techniques:- designed first work on the prototype model of the  Usability Inspection: This is method where original system and test it to check whether thesome developer, user discuss about the different user system is working [1] as per the ser specification orinterface [10]. The commonly used methods are: not. The commonly used technique are: i. Cognitive Walkthrough :It is a i. Coaching method : An expert testerusability inspection method which specifies the is considered who will judge whether the software issequence of steps or actions required by the user as per requirement or not.accomplish a task and the system responses to those ii. Co-discovery method : During aaction. usability test, two test users attempt to perform tasks ii. Heuristic Evaluation : Based on the together while being observed. They are to help eachcurrent requirement of the user inspection is done. It other in the same manner as they would if they werespecifies the involvement of evaluators to judge that working together to accomplish a common goal usingis software is capable of recognizing usability the product. They are encouraged to explain whatprinciples. they are thinking about while working on the tasks. iii. Feature Inspection : It is a lists This technique can be used in the followingsequences of features used to accomplish typical development stages: design, code, test, andtasks, checks for long sequences, cumbersome steps, deployment.steps that would not be natural for users to try, and iii. Performance method: Thissteps that require extensive knowledge/experience in technique is to used to obtain quantitative data aboutorder to assess a proposed feature set. test participants performance when they perform the tasks during usability test. This will generally prohibit an interaction between the participant and 29 All Rights Reserved © 2012 IJARCET
  3. 3. ISSN: 2277 – 9043 International Journal of Advanced Research in Computer Engineering & Technology Volume 1, Issue 2, April 2012the tester during the test that will affect the development stages: design, code, test, andquantitative performance data. It should be conducted deployment.in a formal usability laboratory so that the data can be viii. Think aloud method : Humancollected accurately and possible unexpected factors engineers go to representative userssinterference is minimized. Quantitative data is most workplace and observe them work, to understanduseful in doing comparative testing, or testing against how the users are using the system to accomplishpredefined be marks. their tasks and what kind of mental model the users iv. Question asking protocol : During a have about the system. This method can be used inusability test, besides letting the test users to the test and deployment stages of the development ofverbalize their thoughts as in the thinking aloud the product.protocol, the testers prompt them by asking direct  Usability Inquiry: usability evaluators obtainquestions about the product, in order to understand information about users likes, dislikes, needs, andtheir mental model of the system and the tasks, and understanding of the system by [1] talking to them,where they have trouble in understanding and using observing them using the system in real work (not forthe system. This is a more natural way than the the purpose of usability testing), or letting themthinking-aloud method in letting the test user to answer questions verbally or in written form. Inquiryverbalize their thoughts. methods include: v. Remote testing :. Remote usability i. Field observation : In thistesting is used when tester(s) are separated in space technique, human factors engineers formulateand/or time from the participants. This means that the questions about the product based on the kind oftester(s) cannot observe the testing process directly issues of interest. Then they interview representativeand that the participants are usually not in a formal users to ask them these questions in order to gatherusability laboratory. There are different types of information desired. It is good at obtaining detailedremote testing. One is same-time but different-place, information as well as information that can only bewhere the tester can observe the test users screen obtained from the interactive process between thethrough computer network, and may be able to hear interviewer and the user.what the test user says during the test through speaker ii. Focus group : This is a datatelephone. collecting technique where about 6 to 9 users are vi. Shadow method : If a videotape has brought together to discuss issues relating to thebeen made of a usability test session, the tester(s) can system. A human factors engineer plays the role of acollect more information by reviewing the videotape moderator, who needs to prepare the list of issues totogether with the user participants and asking them be discussed beforehand and seek to gather thequestions regarding their behavior during the test. So needed information from the discussion. This canthis technique should be used along with other capture spontaneous user reactions and ideas thattechniques, especially those where the interaction evolve in the dynamic group process.between the testers and the participants is restricted. iii. Interview : In this technique,But using this technique means that each test takes at human factors engineers formulate questions aboutleast twice as long. Another obvious requirement for the product based on the kind of issues of interest.using this technique is that the users interaction with Then they interview representative users to ask themthe computer needs to be recorded and replayed. these questions in order to gather information vii. Teaching method : During a desired. It is good at obtaining detailed informationusability test, let the test users interact with the as well as information that can only be obtained fromsystem first, so that they get familiar with it and the interactive process between the interviewer andacquire some expertise in accomplishing tasks using the user.the system. Then introduce a naive user to each test iv. Logging actual use : Logginguser. The Novice users are briefed by the tester to involves having the computer automatically collectlimit their active participation and not to become an statistics about the detailed use of the system. It isactive problem-solver. Each test user is asked to useful because it shows how users perform theirexplain to the novice how the system works and actual work and because it is easy to automaticallydemonstrate to him/her a set of pre-determined tasks. collect data from a large number of users workingThis technique can be used for the following under different circumstances. Typically, an interface 30 All Rights Reserved © 2012 IJARCET
  4. 4. ISSN: 2277 – 9043 International Journal of Advanced Research in Computer Engineering & Technology Volume 1, Issue 2, April 2012log will contain statistics about the frequency with software attributes that describe the relative ease forwhich each user has used each feature in the program the users to learn the application.and the frequency with which various events of a). Percentage of Observable Properties: Thisinterest (such as error messages) have occurred. parameter tells whether the software is capable of identifying the observable properties or not. v. Questionnaire : Questionnaires The percentage of observable properties= no ofhave advantages over some other types of surveys in observable properties/ total no of properties.that they are cheap, do not require as much effortfrom the questioner as verbal or telephone surveys,and often have standardized answers that make it b). Type of Interfaces: This tells whether a propersimple to compile data. However, such standardized graphical interface is provided for the better learn-answers may frustrate users. Questionnaires are also ability of the user.sharply limited by the fact that respondents must be iii. Operability: This deals with theable to read the questions and respond to them. Thus, software attributes that are associated to the relativefor some demographic groups conducting a survey by ease of learning the operation of the software.questionnaire may not be practical. a)Complexity of functionality: This tells whether the complexity of the different functionality can affectOut of all this methods we are working with the the software operation or not.usability inquiry method where we will design a b). Ease of use and navigability: This tells us that thequestionnaire system to evaluate usability . navigation to different functionality within the software is easy or not. IV. METRICS UNDER iv. Attractiveness: This describes the CONSIDERATION degree to which the software has been made attractive.The different type of the real time metrics that we a). Use of Graphics: This parameter tells that thewill consider during the usability evaluation methods software is graphically attractive or not and whetherare as given below the graphics are capable of solving the purpose for i. Understandability: This deals with the which they are designed.attribute of software that describe the relative ease of v. Usability compliance: This determinesrecognizing the logical concept and its applicability. whether the software adheres to the compliancea). Documentation: This evaluate that whether the standards of usability or not.documentation are properly written or not. a). Adherence to Compliance Standards: This tells usb). Help system: It tells us that whether a help whether the software is capable of achieving thesupport system is available for the particular software standard been designed.or not.c). Training provided: It deals with whether a proper We have designed a questionnaire based on the abovetraining is provided for the better understandability of metrics. Data will be collect on basis of thisthe software. questionnaire from experts in the specific domain been considered.d). Errors and pop-up: This tells us whether errornotification and pop up are provided for the softwareor not. This help the user to better understand the V. INTREGATING USABILITY INsoftware. SDLC LIFE CYCLEe). International language support: If a software Usability is one of the important phase in anysupport all language then it become very convenient software development process. It is required tofor the user to understand it. This metric measure include it in the SDLC life cycle. Although usabilitywhether this facility is provided or not. is included in SRS documentation where the userf). Online Help Support: This metrics tells us whether specification is considered . If we study the SDLCan online support system is provided for the user life in detail then we will see that usability is beenconvenience or not. done before implementation and after ii. Learn- ability: This deals with the implementation. Before implementation it is done in 31 All Rights Reserved © 2012 IJARCET
  5. 5. ISSN: 2277 – 9043 International Journal of Advanced Research in Computer Engineering & Technology Volume 1, Issue 2, April 2012the feasibility study and after implementation it is implementation phase we evaluate whether thedone at maintenance. In pre-implementation phase software is working and been designed as per userwe deals with the user requirement where the user requirement or not. So based on this concept we havespecification is studied and is tried to be implemented tried to add the usability in SDLC life cycle.in the prototypic model of software. And in post Usability Fig.2 Prototype model including the Usability phase VI. CONCLUSION 7. REFERNCES [1] J. Nielsen "Usability Engineering" p.198, AcademicThis paper has reflected the different type of usability Press, 1993.technique that can be used to calculate the usability [2] P. R. Srivastava, A. P. Singh, K.V. Vageesh,of any software. The paper has given an introduction “Assssment of Software Quality: A Fuzzy Multi –to all the methods that can be employed to calculate Criteria Approach,” Evolution of Computationandthe usability. The technique that we used is very Optimization Algorithms in Softwareeffective since it considered the user specification. EngineeringApplications and Techniques, IGI GlobalUsability can also evaluate the performance too , so USA, 2010, chapter - 11, pp.200-219.can employ this method to evaluate the performance [3] Integrated Software Quality Evaluation: A Fuzzyof any software. As usability is one of the most Multi-Criteria Approach Jagat Sesh Challa, Arindamimportant factor considered in all the Software Paul, Yogesh Dada, Venkatesh Nerella, PraveenQuality Model so it is needed to do a lot of research Ranjan Srivastava and Ajit Pratap Singh Journal ofto provide a method for better estimation of this Information Processing Systems, Vol.7, No.3,parameter. September 2011. [4] http://en.wikipedia.org/wiki/Fuzzy_logic [5] Stephen H. Kan - Metrics & Models in Software Quality Engineering. [6] Lotfi A Zadeh, Fuzzy sets, Information and control, vol 8, pp 338-353, 1965. 32 All Rights Reserved © 2012 IJARCET
  6. 6. ISSN: 2277 – 9043 International Journal of Advanced Research in Computer Engineering & Technology Volume 1, Issue 2, April 2012[7] Lotfi A Zadeh, The concept of a linguisticvariable and its application to approximatereasoning–I, Information Sciences, Volume 8, Issue3, pp 199-249, 1975[8] George K Klier and Tina A Folger, Fuzzy sets,Uncertainity and Information, Prentice Hall, 1988.[9] Fuzzy Logic Based Revised Defect Rating forSoftware Lifecycle Performance Prediction UsingGMR A K Verma, Anil R, Dr. Om Prakash JaiBVICAM’S International Journal Of InformationTechnology Bharati Vidyapeeth’s Institute ofComputer Applications and Management, NewDelhi.[10] C. Wharton et al. "The cognitive walkthroughmethod: a practitioners guide" in J. Nielsen & R.Mack "Usability Inspection Methods" pp. 105-140. Piyush Jain is assistant professor in thedepartment of computer science and engineering in IIMTengineering and technology and pursuing M.Tech from AmityUniversity, Uttar Pradesh. His research area of work is evaluationof software quality. Sanjay Kumar Dubey is Assistant Professorin the Department of Computer Science and Engineering in AmityUniversity Uttar Pradesh, India. His research area includesSoftware Engineering and Usability Engineering.Prof. (Dr.) Ajay Rana is Professor and Director, ATPC, AmityUniversity Uttar Pradesh, India. He is Ph. D. (2005) in ComputerScience and Engineering from Uttar Pradesh Technical University,India. His research area includes Software Engineering andSoftware Testing. He has published more than 40 research papersin reputed National & International Journals. He is the author ofseveral books in the area of computer science. He has receivednumbers ofbest papers/case studies medals and prizes for his work. 33 All Rights Reserved © 2012 IJARCET