* Topic-Similar_BugReports are extracted and input into the Naive
Bayes Multinomial algorithm.
A closer look to the behaviour of both
Bayes Factor estimations (Table 2) reveals that the BIC-based estimation tends to favour the [H.sub.0] against the [H.sub.1] when the p-value calibration finds substantial, strong, very strong or decisive evidences for [H.sub.1].
Some software tools support data mining processes, allowing management of large volumes of data, which is the case of Weka--developed in Waikato in New Zealand and which in this study shows results of the process of the Naive
Bayes and C4.5 algorithms (denominated J48 in the Weka tool).
[[??].sup.2.sub.t(
Bayes)] = 0.0303 + 0.0951[[epsilon].sup.2.sub.t-1] + 0.8854[[sigma].sup.2.sub.t-1] (9)
The first stage is to apply Naive
Bayes algorithm and classifies the preprocessed data sets.
This paper attempted to study the Naive
Bayes algorithm's performance for dataset of different sizes.