Florida Atlantic University
CAP 6673
This assignment involves ordering of software modules, based on the number of faults predicted by a software quality prediction model. Use the predictions obtained with Linear Regression Models we built in Homework I-B. You will use only two of the three models you obtained in Homework I-B. The models to be used for this assignment are
...[Show More]
This assignment involves ordering of software modules, based on the number of faults predicted by a software quality prediction model. Use the predictions obtained with Linear Regression Models we built in Homework I-B. You will use only two of the three models you obtained in Homework I-B. The models to be used for this assignment are as follows: o Linear Regression Model with M5 Method of Attribute Selection: FAULTS = - 0.0516 * NUMUORS + 0.0341 * NUMUANDS - 0.0027 * TOTOTORS - 0.0372 * VG + 0.2119 * NLOGIC + 0.0018 * LOC + 0.005 * ELOC - 0.3091 o Linear Regression Model with Greedy Method of Attribute Selection: FAULTS = - 0.0482 * NUMUORS + 0.0336 * NUMUANDS - 0.0021 * TOTOTORS - 0.0337 * VG + 0.2088 * NLOGIC + 0.0019 * LOC - 0.3255 Obtain the predictions for both the fit data set and the test data set using the above two models. Perform Module Order Modeling for both fit and test data sets using both regression models. Compare the performances of MOM for both linear regression models. Use Alberg Diagram and Peformance Curve for each Model using fit and test data sets. Use tables to summarize the results of MOM. Also provide analysis of your summary. In order to perform the M.O.M. method, the predictions for both fit and test needed to be done. To do this, the first step would use both linear regression models and calculate each module's number of faults. The software used is EXCEL, but another software may be used such task. After obtaining the INTEGER value of the number of faults, application of M.O.M. can be used to rank the prediction. First, the modules of the actual dataset shall be ranked based on highest to lowest of the number of faults it contains. Next, the modules for both prediction models will be ranked from highest to lowest based on their number of faults. After applying a descending order as mentioned, replace both columns with the number of faults of the ACTUAL dataset. Prediction: The actual and predicted number of faults for the fit and test dataset are shown in Figure 2 and 3, respectively: From here, a filter was used for the columns that contained the numeric values of the predicted number of faults. The procedure was then to apply descending order for the number of faults of the actual, M5, and Greedy. The module number is then rearranged to fitt the criteria of the highest number of faults to lowest. The number of faults from the ACTUAL column are to be recorded as the predicted ranking modules for both the M5 and Greedy. Again, the predicted number of faults are not used after ranking has been done and is shown in Figures:
[Show Less]