WisdomModel: convert data into wisdom

Purpose Traditional classification algorithms always have an incorrect prediction. As the misclassification rate increases, usefulness of learning model decreases. This paper presents development a wisdom framework that reduces error to less than 3% without human intervention. Design/methodology/approach The proposed WisdomModel consists four stages: build classifier, isolate misclassified instances, construct automated knowledge base for instances and rectify approach will identify by comparing them against base. If instance is close rule in certain threshold, then this considered misclassified. Findings authors evaluated using different measures such as accuracy, recall, precision, f-measure, receiver operating characteristics (ROC) curve, area under curve (AUC) with various data sets prove its ability generalize involvement. results minimize number at least 70% increase accuracy minimally 7%. Originality/value research focuses on defining practical applications. Despite information system, there still no or algorithm can be used extract from data. general any domain reach wisdom.