Performance-Aware Approach for Software Risk Management Using Random Forest Algorithm

2. Key Demarcation Between Hard And Soft Computing For Bugs Prediction

In case of hard computing based approach, the software bugs prediction is based on the static mathematical formulations while the soft computing based architecture with efficiency and performance aware with the usage of fuzzy based formulations. In these the river formation dynamics, simulated annealing, honey bee approach, random forest or any similar can be used.(Özcan, Ender, Burak Bilgin and Emin Erkan Korkmaz,2008)(Burke E, Kendall G, Newall J, Hart E, Ross P,2003)(Özcan E, Bilgin B, Korkmaz EE,2008)

Figure 1.

Hard computing vs. soft computing

3. Proposed Effective Design And Framework For Software Risk Management

The present work is having key focuses on the development of a novel architecture and implementation using Random Forest Approach so that the higher degree of efficiency and accuracy can be achieved. (Karaboga D, Akay B, 2009) (Karaboga D, Basturk B, 2007) Random Forest algorithmic program may be a supervised classification algorithmic program whereby it creates a forest by a way and makes it random towards the answer or optimization perspectives. There is a direct relationship between the number of trees in the forest and the results it can get: the larger the number of trees, the more accurate the result. But one thing to note is that creating the forest is not the same as constructing the decision with information gain or gain index approach. (Shah-Hosseini H, 2008) (Shah-Hosseini H, 2009) (Glover F, 1997) (Kirkpatrick S, Gelatt CD, Vecchi MP, 1983)

In case of Random forest approach; there is generation of number of decision trees whereby the individual decision is associated with each tree. From all the possible permutations, the best outcome in terms of optimization is achieved. For this approach, the paradigm of voting is done in random forest approach.(Dueck G,1993) (Hansen P, Mladenović N, Perez-Britos D,2001) (Raicu I, Schwiebert L, Fowler S, Gupta SK,2005)

To perform the simulation and implementation, a dataset from the real time software engineers are obtained and extracted so that the training of the bugs based records can be done. The data is inserted for training and further prediction in the following format.

One big advantage of random forest is, that it can be used for both classification and regression problems, which form the majority of current machine learning systems. I will talk about random forest in classification, since classification is sometimes considered the building block of machine learning. Below you can see how a random forest would look like with two trees. (Goldberg DE, Holland JH, 1998)(Kaur R, Dhindsa KS, 2014) (Kaur S, Dhindsa KS, 2014)(Chawla A, Dhindsa KS, 2014) (Fister Jr, I., Yang, X.S., Fister, I., Brest, J., Fister, D, 2013) (Goyal L. M., M. Mittal and Sethi, J. K, 2016)

Figure 2.

Features identification in random forest