International Journal of Engineering and Technology

Missing data occur in almost all serious statistical analyses. In statistics, imputation is the process of replacing missing data with substituted values. Simple imputation is attractive often used to impute missing data whereas multiple imputation generates right value to replace. This paper evaluates multiple imputation of missing data in large datasets and the presentation of MI focuses on several unsupervised ML algorithms like mean, median, standard deviation and Supervised ML techniques for probabilistic algorithm like NBI classifier. This survey carried out using comprehensive range of databases, for which missing cases are first filled by several sets of reasonable values to create multiple finalized datasets, then standard complete data procedures are register to each finalized dataset, and eventually the multiple sets of results are merge to produce a single inference. Main goal is to provide general guidelines on selection of suitable data imputation algorithms and also implementing Bolzano theorem in machine learning techniques to evaluate the performance of every sequence of rational and irrational number has a monotonic subsequence. To evaluate imputation of missing data, the standard machine learning repository dataset has been used. Experimental results shows the proposed approach have good accuracy and the accuracy measured in terms of percentage.