Test Case Prioritization Using Deep Learning Hybrid Approach

Exhaustive testing and coverage are not the effective method to perform the validation in manual execution, automation and continuous integration and development environment. In the current mode of work, we encountered backlog of execution tasks up to three weeks instead of planned one week duration for completion. Therefore, this research will prioritize on test case prioritization which allows validation engineer and automation framework to execute the validation in operative manner.

To ensure the validation effectiveness of reducing redundancy of the test cases, a deep learning method and greedy approach on selecting the best combination of test cases which have complete validation coverage will be proposed. The reduction will impact significantly on the project cost in term of resource catered.

These results suggested that using hybrid approach will expedite the progress approximately 20%. From organization point of view, this proficiency results in cost saving and having quality products to consumer to use. - from test case prioritization perspective, audience will learn on how to implement test case redundancy method, for instance Greedy approach maximizing the test case coverage with the minimum number of test cases.
- expose the usage of the deep learn

Willam Loo, Validation Engineer, Intel Microelectronics Sdn Bhd

I come from Penang, Malaysia and am currently working in Intel Penang. I have been working with Intel for 6 years and previously I had worked at AMD about 1 year. Currently I'm taking master degree at Universiti Malaysia Pahang (UMP) in Software Engineering and my research area is on the validation methodology improvement. 

Kin Choy Yip, Principal Engineer, Intel Microelectronics Sdn Bhd

Kin Choy Yip works with Willam and has assisted him throughout his career at Intel.