Predicting National Team Rank in Asian Game Using Model Tree
Many people are interested in predicting the outcome of sporting contests. However, one of the reasons that sport attracts so much attention is that the outcome of a contest is not perfectly predictable. In this paper we tried to predict the success of nations at the Asian Games through macro-economic, political, social and cultural variables. we used the information of variables include urban population, Education Expenditures, Age Structure, GDP Real Growth Rate, GDP Per Capita, Unemployment Rate, Population, Inflation Average, current account balance, life expectancy at birth and Merchandise Trade for all of the participating countries in Asian Games from 1970 to 2006 in order to build the model and then this model was tested by the information of variables in 2010. The prediction is based on the number of golden medals acquired each country. In this research we used WEKA software that is a popular suite of machine learning software written in Java. Japans’s stability is entirely consistent with it’s variables in all of the courses held. The value of correlation coefficient between the predicted and original ranks is 75.5%. We tried to design the pattern that:
To improve sport in each country and get the better international ranks according to it’s facilities, potential sources and the comparison with other countries. Managers and planners take the appropriate policies and determine long-term, middle-term and short-term goals in sport according to political, cultural, economic and social factors.
Keywords: Prediction, Asian Game, Macro variable and Model Tree