Enhanced K-Means Clustering Algorithm for Feasibility Assessment of ACC

摘要

Contracts formulation with a minimalistic intervention by developers is crucial to methodically bring about optimized contracts providing extended and enhanced software testing. The proposed scheme to derive the contracts in an automated mode encompasses of shaping the behavior related and structural dependency details as constraints on a Decision tree and subsequently concretize them as Automated Code Contracts (ACC). A Contract refinement is thereafter pursued to harness reinforced feasible contracts. This is performed by tailoring the K-means Clustering algorithm using Neural Networks related Activation functions to construct precise clusters of analogous contracts and at the same time seek finer Computational and memory performances with efficiency gains over bug detecting ability.