Missing Data: Its Emergence in the Real-world-A Practical Review on Google Play Apps dataset using Python

Abstract

In the present up-to-date analysis world, the usage of knowledge (data) is increasing day by day. The unprocurable data is referred to as missing data and also the attainability of dirty data results in the production of ambiguous results or wrong predictions. Missing data earned the researchers' interest because of its surprising emergence throughout experimentations. Varied researchers projected many techniques to handle missing or incomplete data. This paper majorly focuses on missing data and its emergence throughout experimentations. The perception, application, benefits, and downsides of the approaches are mused. As a part of experimentation, this paper conducted varied experiments and reviews varied imputation techniques such as Complete Case Analysis (CCA), handling missing numerical data, and handling missing categorical data to impute missing or incomplete data on an open-source data set namely Google play store apps data. The experiments are conducted on the Anaconda Jupyter notebook using Python programming. We conducted the experiments before imputing the missing values and after imputing the missing values. The experimental studies show the requirement of handling missing data.