engleski

Analyzing growth: A data-driven approach

Aim of this study was to provide a data-driven methodology and framework to auxology, in order for researchers and medical doctors to benefit from large data repositories, from which it is possible to extract hidden, meaningful - but often counterintuitive or unpredicted– information, or to corroborate current knowledge. A large data set of 41, 085 young athletes both genders, aged 7-17, engaged in chosen individual and team sport disciplines, containing clinical and anthropometric data, was investigated by means of traditional statistical methods and innovative data mining/machine learning algorithms. In particular, classification algorithms like decision tree have been adopted in order to make predictions on different biomedical variables. Data mining correctly classified individuals in certain age groups, of both genders, who are engaged in two categories of sports. In case of large samples, data mining has proved to be an effective predictive and decision support tool, which can be used to improve data analysis, beside traditional statistical techniques.

data mining; forecasting

nije evidentirano