Building Defect Prediction Models by Online Learning Considering Defect Overlooking

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Fedorov, Nikolay, Yamasaki, Yuta, Tsunoda, Masateru, Monden, Akito, Tahir, Amjed, Bennin, Kwabena Ebo, Toda, Koji, Nakasai, Keitaro
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929316755931136
author Fedorov, Nikolay
Yamasaki, Yuta
Tsunoda, Masateru
Monden, Akito
Tahir, Amjed
Bennin, Kwabena Ebo
Toda, Koji
Nakasai, Keitaro
author_facet Fedorov, Nikolay
Yamasaki, Yuta
Tsunoda, Masateru
Monden, Akito
Tahir, Amjed
Bennin, Kwabena Ebo
Toda, Koji
Nakasai, Keitaro
contents Building defect prediction models based on online learning can enhance prediction accuracy. It continuously rebuilds a new prediction model, when a new data point is added. However, a module predicted as "non-defective" can result in fewer test cases for such modules. Thus, a defective module can be overlooked during testing. The erroneous test results are used as learning data by online learning, which could negatively affect prediction accuracy. To suppress the negative influence, we propose to apply a method that fixes the prediction as positive during the initial stage of online learning. Additionally, we improved the method to consider the probability of the overlooking. In our experiment, we demonstrate this negative influence on prediction accuracy, and the effectiveness of our approach. The results show that our approach did not negatively affect AUC but significantly improved recall.
format Preprint
id arxiv_https___arxiv_org_abs_2404_11033
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Building Defect Prediction Models by Online Learning Considering Defect Overlooking
Fedorov, Nikolay
Yamasaki, Yuta
Tsunoda, Masateru
Monden, Akito
Tahir, Amjed
Bennin, Kwabena Ebo
Toda, Koji
Nakasai, Keitaro
Software Engineering
Building defect prediction models based on online learning can enhance prediction accuracy. It continuously rebuilds a new prediction model, when a new data point is added. However, a module predicted as "non-defective" can result in fewer test cases for such modules. Thus, a defective module can be overlooked during testing. The erroneous test results are used as learning data by online learning, which could negatively affect prediction accuracy. To suppress the negative influence, we propose to apply a method that fixes the prediction as positive during the initial stage of online learning. Additionally, we improved the method to consider the probability of the overlooking. In our experiment, we demonstrate this negative influence on prediction accuracy, and the effectiveness of our approach. The results show that our approach did not negatively affect AUC but significantly improved recall.
title Building Defect Prediction Models by Online Learning Considering Defect Overlooking
topic Software Engineering
url https://arxiv.org/abs/2404.11033