Correlation Coefficient Definition

The strength of linear association among two variables will be calculated. Most of the time the correlation will be between -1.0 and +1.0. The positive relationship will be obtained in cases of positive correlation. The negative relationship will be obtained in cases of negative correlation.

FORMULA

Correlation Coefficient

Where

ZY2 = Sum of square Second Scores.

LX2 = Sum of square First Scores.

ZY = Sum of Second Scores.

ZX = Sum of First Scores.

ZXY = Sum of the product of First and Second Scores.

Y = Second Score.

X = First Score.

N = Number of values or elements.

Empirical Validation

Here we calculate the Component Selection Efforts by two methods: one is fuzzy logic-based and another is the Analytical Hierarchical Process (AHP). Finally, we calculate the correlation coefficient between these two outputs and the value of this comes out at 0.714. Hence, we conclude that the proposed technology is valid. Validation is shown in Table 4.7.

TABLE 4.7

Correlation between Selection Efforts Using Fuzzy Rule and Using AHP

Selection Efforts Using Fuzzy Rule Base (X)

Selection Efforts Using AHP (Y)

Correlation coefficient between X and Y

0.51

0.425

0.714

0.713

0.8160

0.51

0.48

0.51

0.3162

0.572

0.475

0.718

0.653

0.713

0.572

0.361

0.243

Exercise

Case Study 1

There is a food service provider, “Blue Pluto”, which provides catering services for various occasions such as family functions, office parties, wedding parties, birthday parties, etc. It provides catering services either at the client’s personal venue or in the venues with which there is a tie-up with Blue Pluto. The service offers different packages and also the client may customize based on their requirements. Catering is a multifaceted segment of the food service industry. Blue Pluto has many sites in different cities with some different services but presently they all work manually. Since Blue Pluto is a large venture it requires a software application to maintain master and transaction records and automate its system. Initially Blue Pluto wants to automate only one site. Along with automating basic functionalities, there are the following non-functional requirements with different weights too:

Reusability - Low.

Portability - High.

Functionality - Very High.

Security - High.

Performance - High.

Use the fuzzy-based effort estimation technique to estimate the effort required to develop the software

Reusability

Portability

Functionality

Security

Performance

0.65

0.5

0.8

0.45

0.4

Fake Feedback Identification System (FFIS)

FIGURE 4.10 Fake Feedback Identification System (FFIS).

Case Study 2

There is a fake feedback identification system application which identifies the fakeness of feedbacks collected through various resources. This is a three-level system. The first level is “flip any two questions”, the second level is “select answer from drop down box for two questions”, and the third level is “preparing fill in the blank for two questions”. The answers filled in by candidates the first time are compared with the answers received through these three levels and then with an internal computation feedback is marked as fake or fair. Based on the different environment systems can be modified and applied. For example, it could be used in colleges, universities, hotels, hospitals, online shopping, etc.

The framework of the system can be understood through Figure 4.10.

There are two proposals given by a development company.

A: Along with all the basic functionalities shown in the figure, the application will be very highly reusable and portable while the security, functionality, and functionality are at a low level.

B: Along with all the basic functionalities shown in the figure, in the application reusability, portability, security, and functionality are at high.

Which proposal will be the cheaper one if we use the fuzzy logic effort estimation approach discussed in the chapter and the weights of all the quality parameters are 0.25 for the Fake Feedback Identification System (FFIS)?

References

Carrasco-Colomo-Palacios, R.. Lopez-Cuadrado, J. L., & Penalvo. F. J. G. (2012). SEffEst: Effort estimation in software projects using fuzzy logic and neural networks. International Journal of Computational Intelligence Systems, 5(4), 679-699.

Martin, C.L., Pasquier, J.L., Yanez, C.M., Tornes, A.G. (2005). Software development effort estimation using fuzzy logic: A case study. Proceedings of the sixth Mexican International Conference on Computer Science (ENC’05), Puebla, Mexico.

Parkash, K., Mittal, H., & Mittal, H. (2010). Software cost estimation using fuzzy logic. ACM SICSOFT Software Engineering, 55(1), 1-7.

Patil, L. V., Shivale, N.M., Joshi, S.D., Khanna, V. (2014). Improving the accuracy of CBSD effort estimation using fuzzy logic. IEEE International Advance Computing Conference, IACC, Guragon, India.

Saaty, T. L. (1994). How to make a decision: The analytic hierarchy process. Interfaces, 24(6), 19-43. doi:10.1287/inte.24.6.19.

Seth, K., Sharma, A. Seth, A. (2009).Component selection efforts estimation-a fuzzy logic based approach. International Journal of Computer Science and Security,(IJCSS), 3(3), 210-215.

Sharma, A.. Grover, P. S., & Kumar, R. (2009). Reusability assessment for software components. ACM SICSOFT Software Engineering Notes, 39(2): 1-6.

Xu, Zhiwei, & Khoshgoftaar, Taghi M. (2004). Identification of fuzzy models of software cost estimation. International Journal of Fuzzy Sets and Systems, /95(1),141-163. Elsevier.

Zadeh, L. A. (1965). Fuzzy sets. Information and Control, 8(3), 338-353.

 
Source
< Prev   CONTENTS   Source   Next >