Abstract

One of the most exciting outdoor games that reached everyone heart is cricket. There are several series held and one among that created a magnificent history in the arena of sports is Indian Premier League (IPL). It has reached its popularity with successful brand in the world of sports and usually will be conducted among 8 teams. This proposed paper is specifically concentrating on enactment and measuring the difference between the models to foretell the captivating team of an IPL match. Data is accessed by the computer programs developed using Machine learning to build models. As of now, data analysis is need for each and every fields to examine the sets of data to extract the useful information from it and to draw conclusion and as well make decisions according to the information. The algorithm first analyses the data to create a model, specifically for understanding the patterns or trends. For creating the mining model, the model is optimized by selecting parameters and iterating. To extract actionable patterns and detailed statistics, the parameters are then fed into the dataset. This work focuses on finding the meaningful information about the IPL Teams by using the functions of R Package. R reduces the complexity of data analysis as it displays the analysis results in the form of visual representations. The dataset is loaded and a set of pre-processing is done followed by feature selection. Four machine learning algorithms Decision Tree, Naive Bayes, K-Nearest Neighbour and Random Forest are applied and the results are compared to measure the accuracy, precision, recall and sensitivity. The best of the four machine learning techniques is then applied to predict the winner and visualizes the results as graphs.