Recently Published
Terrorism in the World - 1970 to 2017
Evolution of terrorism in the world - 1970 to 2017
Terrorism in the World - 1970 to 2017
Visualizing the evolution of terrorism in the world from 1970 to 2017
Deconstruct, Reconstruct Web Report - Share of Renewable Energy in Total Energy Produced by North American States
The objective of this report was to deconstruct a problematic data visualisation and ultimately reconstruct it with the raw data, rectifying the issues observed.
Body Mass Index vs. Success at Olympics
The dataset "athlete", which contains information about athletes participating at Olympics from the year 1896 to 2016, was merged with the other dataset "noc" which contains country names for the respective NOC codes. The resulting dataset had 271116 observations across 16 variables about the athlete, with “Age”, ”Weight”, “Height” and “Medal” being the most important ones. The pre-processing of data was carried out with the aim of building a dataset that helps find relation between the Body Mass Index of an athlete and his/her success at the Olympics. The data structures of variables of the dataset were inspected and a few variables were converted into factor format. Body Mass Index for each athlete was calculated by using “Height” and “Weight” variables of the dataset and adding the result under a new variable “BMI”. The resulting dataset was scanned for missing values, which were handled by imputation followed by omission. Outlier scan detected numerous outliers, which were not manipulated as the data was about the physical attributes and performance of individual athletes at the Olympics. However, the critical BMI values, falling out of (15,60) range were filtered out to maintain the integrity and relevance of the dataset. Finally, the “Weight” variable’s values were found to be right-skewed, which were transformed into a more symmetrical, normally distributed graph, by using cube-root transformation.