Predictive Modelling and Analytics on Big Data for Diabetic Management

Nowadays Analytics are embedded into our daily life. A large number of analytical models are developed so far to enhance the qualitative relationships among the covariates. When modelling features are small, a mathematical model itself describes the relationship among the elements of the domain. Research suggests that a simple algorithm without convincing the performance and accuracy of the statistical analysis with a large volume of data is more accurate than a sophisticated algorithm with little data. A Diabetes data set from 1999 to 2008 with 101676 distinct records were taken and statistical techniques were applied to the data sets, and we noted that both interpretability and justifiability are balanced against statistical performance. This analysis of a large clinical database assessment informs the future directions which might lead to improvements in patient safety.