A risk-focused levels classification based on test case generation and prioritization by using historical system failure data

Name: Rachanon Aiemkrong

Organization : Rajamangala University of Technology Thanyaburi. Faculty of Science and Technology

Email : rachanon_a@mail.rmutt.ac.th

Name: Nipat Jongsawat

Organization : Rajamangala University of Technology Thanyaburi. Faculty of Science and Technology

Email : nipat_j@rmutt.ac.th

Name: Wiparat Busyatras

Organization : Rajamangala University of Technology Thanyaburi. Faculty of Science and Technology

Email : wiparat_b@rmutt.ac.th

LCSH: Computer software -- Quality control -- Data processing

LCSH: Software failures -- Classification

LCSH: Risk assessment -- Mathematical models

LCSH: Supervised learning (Machine learning)

LCSH: Decision trees

Abstract: In software case tracking systems, risk level serves as a key criterion for prioritizing and addressing defect cases. Therefore, this study employs a dataset consisting of log activity records for tracking problem resolution processes. This Research proposes a model to predict risk levels without depending exclusively on numerical value-based calculations. The dataset consists of ten features as timestamp, Module, Module_Type, Severity, Description, Problem_Category, Root_Cause, Related_test_case_id, Status, and Fix_Time_Hours. This work proposes a supervised learning approach to predict the risk level. The labels of the risk levels include Low, Medium, High, and Very High, for log-based case resolution activities. The comparative model includes nine algorithms as Naive Bayes, Generalized Linear Model, Logistic Regression, Fast Large Margin, Deep Learning, Decision Tree, Random Forest, Gradient Boosted Trees, and Support Vector Machine (SVM). The high-performance results indicate that Deep Learning, Decision Tree, Random Forest, and Gradient Boosted Trees achieved accuracy rates of 89 percent, 92 percent, 91 percent, and 90 percent, respectively. Among these, the Decision Tree algorithm achieved the highest accuracy percentage. These results demonstrate that risk level prediction can be improved the problem resolution processing. When the risk level of a case is identified in advance, appropriate attention and resources can be allocated more effectively to optimize case management.

King Mongkut's University of Technology North Bangkok. Central Library

Address: BANGKOK

Email: library@kmutnb.ac.th

Created: 2026

Modified: 2026-08-11

Issued: 2026-08-11

บทความ/Article

application/pdf

BibliograpyCitation : In ECTI Association and Burapha University. Faculty of Engineering. 2026 23rd International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology (ECTI-CON 2026) (pp.493-496). Piscataway, NJ : IEEE

eng

©copyrights King Mongkut's University of Technology North Bangkok

ลำดับที่.	ชื่อแฟ้มข้อมูล	ขนาดแฟ้มข้อมูล	จำนวนเข้าถึง	วัน-เวลาเข้าถึงล่าสุด
1	ECTI-CON 2026pp.493-496.pdf	1.08 MB

ใช้เวลา

0.03029 วินาที