Correcting Missing Data Anomalies with Clausal Defeasible Logic

Databases are used globally to store essential information required for various business applications such as automated data capturing. Unfortunately, due to missing record anomalies present within the repository, the overall integrity of stored information is compromised. Currently, filtration and rule-based techniques have been proposed to correct the problem, but due to a lack of high-level reasoning, ambiguous scenarios lead to anomalies persisting within the database. In this paper, we propose an enhanced Non-Monotonic Reasoning cleaning architecture that utilises intelligent analysis coupled with Clausal Defeasible Logic to rectify the missing data by generating and restoring imputed data. From our experimental evaluation, we have found that our proposed technique surpasses other leading intelligence classifiers such as Bayesian and Neural Networks.