1. Home »
  2. Supervisor: Dr. Tamim Ahmed Khan
Supervisor: Dr. Tamim Ahmed Khan

 

Test Case Generation and Automated Execution Using Large Language Models

 

Supervisor: Dr. Tamim Ahmed Khan, Professor

Campus/School/Dept: BUIC H-11/BSEAS/SE

Supervisory Record:  

  • PhD Produced: 4
  • PhD Enrolled: 2
  • MS/MPhil Produced: 57
  • MS/MPhil Enrolled: 3

 

Topic Brief Description:

Software fault prediction utilizes techniques such as data mining, statistical analysis, machine learning, and software metrics to anticipate potential faults in software systems before they arise. The (SDP-BB) model integrates Bidirectional LSTM (BiLSTM), BERT, and data augmentation methods to analyze code context in a bidirectional manner, capture semantic meaning, and enhance data generation from existing code. There is a technology gap due to emergence of AI we are going to introduce an explainable AI feature by using large language models through which software faults would predict, test cases generated and at last it would explain the whole situation.

 

Research Objectives/Deliverables:

  1. Develop AI models to automate the identification of faulty code considering metrics such as C&K metrics.
  2. AI models based generation of test cases, test coverage, and test oracles containing test data in input/output pairs
  3. Evaluate the performance of AI models against previously developed Deep learning models.
  4. Investigate the interpretability and explainability of AI-driven test case generation and execution.

 

Research Questions: 

  1. How can we develop AI models to automate the identification of faulty code considering metrics such as C&K metrics?
  2. How can we train AI agents using LLMs for generation of test cases, test coverage, and test oracles containing test data in input/output pairs accurately?
  3. How can we evaluate the performance of AI models and compare with previously developed Deep learning models?
  4. How can we augment AI driven test case generation using interpretability and explainability?

 

Candidate’s Eligibility Profile:

  1. The applicant must have an MS/MPhil/Equivalent degree in electrical engineering with CGPA >0. Besides, applicants must have a strong background in mathematics, programming, optimization theory and related fields.
  2. Experience with programming languages such as Python. Candidates should thrive in an international environment and have excellent communication skills to actively contribute to team research efforts.
  3. Proficiency in spoken and written English is essential. We value independence and responsibility while promoting teamwork and collaboration among colleagues.