Stat 504 Analytics - Statistical Learning and Predictive Modeling

Professor: Stephen Lee

30 November 2016

Abdullah Alowairdhi

University Of Idaho, Moscow, ID

Contents:

  • Projects II & III

  • Data Visualization D3.js

  • Data Science (the Future of the Internet)

Project II

Analysis of Soccer's Player Performance

Identifying the most important attributes of Soccer player's performance which determine their overall ratings by using series of supervised and unsupervised classifications and regression techniques

Reaserch Question

“Does the dataset tell us what are the set of skills/attributes that determines the overall rating of a soccer plyer performance? Therefore, we can decide to buy that player or not.”

Reaserch Hypotheses

“Soccer players with certain set of skills/attributes are more likely to have higher performance rating than other players”.

The Data Set

Players and Teams' attributes sourced from EA Sports' FIFA video game series, http://sofifa.com/ FIFA series and all FIFA assets property of EA Sports.

  • 7 datasets - 183 attributes & 200,000 observations

  • 2 dataset Selected :

  • Player dataset with 11,060 obs and 7 attrbs &

  • Player_Attributes dataset with 183,978 obs and 42 attrbs

Descriptive and Predictive analysis

.