Academia.eduAcademia.edu

Figure 7 – uploaded by Amelia Ritahani Ismail

See full PDFdownloadDownload figure
TABLE VII. COMPARISON OF MACHINE LEARNING MODELS WITH KAUSHIK ET.AL, 2012

Table 7 VII. COMPARISON OF MACHINE LEARNING MODELS WITH KAUSHIK ET.AL, 2012

Related Figures (14)

TABLE I. INTERMEDIATE COCOMO DRIVERS
TABLE I. INTERMEDIATE COCOMO DRIVERS
TABLE I. DATASET USED IN THE EXPERIMENT
TABLE I. DATASET USED IN THE EXPERIMENT
The data is pre-processed in order to calculate the effort estimation. In this experiment, the data is imported in r studio. Mice package is used to check the missing values, the datasets contain no missing values. The value of the drivers is in numerical weight converted to numerical values due to avoid bias during constructing the machine learning model. The mode constant is assigned based on the COCOMO predefined values.
The data is pre-processed in order to calculate the effort estimation. In this experiment, the data is imported in r studio. Mice package is used to check the missing values, the datasets contain no missing values. The value of the drivers is in numerical weight converted to numerical values due to avoid bias during constructing the machine learning model. The mode constant is assigned based on the COCOMO predefined values.
Fig. 3. Correlation of COCOMO NASA 1 Attributes. Fig. 2. Correlation of COCOMO81 Attributes.
Fig. 3. Correlation of COCOMO NASA 1 Attributes. Fig. 2. Correlation of COCOMO81 Attributes.
Then, the project builds predictive machine learning models with COCOMO81, COCOMO NASA 1, COCOMO NASA 2 and Kaushik et al datasets, using all attributes. The predictive machine learning models are Support Vector Machine, Linear Regression, Regression Tree and Random forest. The result record is in the Table IV. The project evaluates the result and records in a table.
Then, the project builds predictive machine learning models with COCOMO81, COCOMO NASA 1, COCOMO NASA 2 and Kaushik et al datasets, using all attributes. The predictive machine learning models are Support Vector Machine, Linear Regression, Regression Tree and Random forest. The result record is in the Table IV. The project evaluates the result and records in a table.
Fig. 4. Correlation of COCOMO NASA 2 Attributes.
Fig. 4. Correlation of COCOMO NASA 2 Attributes.
Table IV explains the experiment is made on COCOMO81 with training dataset of 80 percent and the testing dataset is 20 percent. Based on the Table IV, Support Vector Machine, Regression Tree and Random Forest are significant due to the p-value which is less than 0.005 except Linear Regression model which p-value is 0.03254. TABLEIV. COMPARISON OF MACHINE LEARNING MODELS WITH COCOMO81 DATASET TABLE V. COMPARISON OF MACHINE LEARNING MODELS WITH COCOMO NASA | DATASET TABLE VI. COMPARISON OF MACHINE LEARNING MODELS WITH COCOMC NASA 2 DATASET
Table IV explains the experiment is made on COCOMO81 with training dataset of 80 percent and the testing dataset is 20 percent. Based on the Table IV, Support Vector Machine, Regression Tree and Random Forest are significant due to the p-value which is less than 0.005 except Linear Regression model which p-value is 0.03254. TABLEIV. COMPARISON OF MACHINE LEARNING MODELS WITH COCOMO81 DATASET TABLE V. COMPARISON OF MACHINE LEARNING MODELS WITH COCOMO NASA | DATASET TABLE VI. COMPARISON OF MACHINE LEARNING MODELS WITH COCOMC NASA 2 DATASET
In this research further experiment is to analysis the regression models with selected attributes COCOMO dataset using A measure and Wilcoxon Rank Sum test.
In this research further experiment is to analysis the regression models with selected attributes COCOMO dataset using A measure and Wilcoxon Rank Sum test.
TABLE VIII. COMPARISON OF COCOMO NASA 1| DATASET WITH DIFFERENT ATTRIBUTES
TABLE VIII. COMPARISON OF COCOMO NASA 1| DATASET WITH DIFFERENT ATTRIBUTES
Fig. 5. Comparison of Actual Effort and Estimated Effort. From the Fig. 5, the machine learning model support vector machine and random forest show similar pattern towards to the actual effort, while the calculation of COCOMO deviates at 100, 14, 302,113, 350 and 339 lines of codes. Support Vector Machine and Random Forest has proximity predicted effort values compared to effort prediction of COCOMO models. Refer to appendices for development of machine learning model.
Fig. 5. Comparison of Actual Effort and Estimated Effort. From the Fig. 5, the machine learning model support vector machine and random forest show similar pattern towards to the actual effort, while the calculation of COCOMO deviates at 100, 14, 302,113, 350 and 339 lines of codes. Support Vector Machine and Random Forest has proximity predicted effort values compared to effort prediction of COCOMO models. Refer to appendices for development of machine learning model.
enough evidences to support null hypothesis and reject the alternative hypothesis since the p-value of both machine learning models is higher than 5 percent. TABLE IX. COMPARISON OF COCOMO NASA 2 WITH DIFFERENT ATTRIBUTES In the experiment of machine learning models are using all attributes, there are large differences result obtained from Support Vector Machine models and Random Forest model. The correlation accuracy of the Random Forest is lower than Support Vector Machine however Random Forest has higher MMRE value compared to Support Vector Machine. Moreover, Vargha and Delaney A measure, Support Vector Machine and large difference between the predicted and actual effort values, while Random Forest has closer accuracy of control and experimental sample. The Wilcoxon rank sum test for Support Vector Machine and Random Forest model are Statistically significant where there are enough evidence to support null hypothesis and reject the alternative hypothesis, as the p-value of both machine learning is higher than 5 percent.
enough evidences to support null hypothesis and reject the alternative hypothesis since the p-value of both machine learning models is higher than 5 percent. TABLE IX. COMPARISON OF COCOMO NASA 2 WITH DIFFERENT ATTRIBUTES In the experiment of machine learning models are using all attributes, there are large differences result obtained from Support Vector Machine models and Random Forest model. The correlation accuracy of the Random Forest is lower than Support Vector Machine however Random Forest has higher MMRE value compared to Support Vector Machine. Moreover, Vargha and Delaney A measure, Support Vector Machine and large difference between the predicted and actual effort values, while Random Forest has closer accuracy of control and experimental sample. The Wilcoxon rank sum test for Support Vector Machine and Random Forest model are Statistically significant where there are enough evidence to support null hypothesis and reject the alternative hypothesis, as the p-value of both machine learning is higher than 5 percent.
Related topics:
Computer ScienceMachine Learning
580 California St., Suite 400
San Francisco, CA, 94104
© 2025 Academia. All rights reserved