A Novel Two Tier Missing at Random Type Missing Data Imputation using Enhanced Linear Interpolation Technique on Internet of Medical Things

Objectives: Data collection and distribution are essential components required for the victory of Internet Medical Things (IoMT) system. Generally, missing data is most recurrent problem that impacts an overall system performance. Methods: Missing in IoMT systems can be caused by various factors, including faulty connections, external attacks, or sensing errors. Although ubiquitous IoT, imputation hardly ever observed setting. As a result, doing analytics on with values causes deterioration accuracy dependability analysis outputs. To achieve excellent performance, must imputed once it occurs such systems. Therefore, this paper proposes novel Two Tier Imputation (TT-MDI) technique at random (MAR) type using enhanced linear interpolation technique. Findings: The proposed TT-MDI algorithm has two tiers imputing MAR was tested Kaggle Machine Learning Repository’s cStick dataset. Utilizing distances between class centroids their related instances, first tier aims to identify threshold. identified threshold then used second impute data. According experimental findings, work offers higher accuracy, precision, recall, F-measure dataset than included when compared original Novelty: consists tiers. uses Manhattan instances discover Next, discovered Enhanced Linear Interpolation Keywords: Things; Data; Threshold Discovery; Distance