Bug Prediction Problem Statement 2: Bug Prediction Application Bug Prediction and Identification is a key activity for software maintenance. The measures on lines of code, McCabe metrics, Halstead measures, branch-count are identified to decide a potential bug. Recently, machine learning models are being leveraged to assess code due to their ability to learn from large number of parameters (features) and faster decision making. Lately, importance of machine learning models explainability has become imminent. This is more important as any wrong decision making can lead to wrong prediction. As part of this hackathon, we need you to build a BugPrediction software application that gives flexibility to technology personnel to define parameters (features) for any code with weightages.
These features should only be numeric. This application is to be deployed as an API for modularity, scalability and management of user roles. A technology person can access code’s data of a particular software product that he/she has access and perform rule and machine learning based assessments. The technology person may also seek explainability of the recommendation made by machine learning to make the final decision. The decision of the technology person to decide whether a bug is present or not has to be provided as a feedback to BugPred so that training data for ML model is enhanced.
Dataset – http://promise.site.uottawa.ca/SERepository/datasets/cm1.arff (Links to an external site.)
Keywords – API, MLOps, Usage of rule based and ML models, Visualization, Explainability