Ipl-Kaggle-Dataset-Analysis - View
Being a sports aficionado, I was always interested to work with a sports dataset. We have good knowledge about cricket and thus, we thought the IPL dataset would be a great to generate some meaningful insights. We were always curious on what factors are players purchased in auctions and wanted to kind of decipher the entire decision process.
It gives teams in-depth analysis using visualization tools in JavaScript to understand what factors contribute in a champion team. Uses the XGBoost Machine Learning technique to predict young players to watch out for in the league and the valuable experienced players.
We got the open data-set from Kaggle and did some data-preprocessing using Numpy and Pandas. New columns and derived data-sets were created using mathematical computations in Python. These plot values were passed as a list to the visualization in JavaScript using HighCharts in the Angular framework.
The data-set wasn't big enough to train a Machine Learning Model, so we had to derive features from the data we had. Figuring out the different visualization tools and choosing the best one.
Winning 2nd Runners-Up at HackDFW for this.
It can be used in actual auctions and the accuracy could be increased with more data and features. These techniques can also be applied to sports fields like NBA, NFL and others.