Sports Analytics

Embed Size (px)

Citation preview

Sports Analytics: Designing a Tennis Game(Singles) Analysis Decision Support Using Big Data

Sports Analytics: Designing a Tennis Game(Singles) Analysis Decision Support Using Big Data

Introduction:The amount of data continues to grow at enormous rate even through the data stores are already vast.The primary challenges areto make the database a competitive business advantage by converting raw data into useful information.

What is DATA mining:Data mining refers to extracting or mining knowledge from large amounts of data.

Many people treat Data mining as a KDD Knowledge Discovery in DB

KD - science of finding patterns within sets of data.

Advantages of Data MiningProvides new knowledge from existing dataPublic databasesGovernment sourcesCompany Databases

Old data can be used to develop new knowledge

Objective:The Objective of this project is: to develop tennis game in javagenerate a computer based analysis tool to analyze wins and losses. provide a meaningful recommendation to improve the indivisible performance in a tennis game.

Using the players history and current situation we are going to predict

Player A or Player B probability of winning game;Suggestions to win a game against opponent; Suggestion to improve from errors

Real time Example:IPL RCB v/s MIRCB needs 160 runs to win from 20 overs with RR of 8.KEY point to win for RCBGAYLE have to score 50+ runs.RCB won all matches when GAYLE scored 50+ runs.STARC have to have minimum economy of 3

KEY point to win for MI

MALINGA have to take 2+ wickets.MI have to take at least 1 wicket in 5 overs.

Modules:Modules used in GAME DEVELOPMENTPlayer RegistrationTournament registrationPlayer selectionGame playing

Player registration:

Tournament Enrolment:

Player Selection:Player can be selected from multiple players for the tournament

Methodology The player has to register firstAll his current and historical data will be available in the databaseThe current situation of the game and his historical data, software is going to predictWinning probability of the playerSuggestions to win a game against opponentSuggestion to improve from errors

GAME DEVELOPMENT

Data mining technique:ClassificationPrediction

Classification

Classification: Given a set of items that have several classes, and given the past instances (training instances) with their associated class, Classification is the process of predicting the class of a new item.identify to which class it belongsEx: Finding player grades.

Technique for ClassificationDecision-Tree ClassifiersAdaptive Boosting

Prediction : It is used to predict missing or unavailable numerical data values rather than class labels

Ex: prediction of winning.

Some commercial toolsOracle Data MinerData To KnowladgeSASClementineEtcConclusionData mining is a decision support process in which we search for patterns of information in data.Overlaps with machine learning , statistics, artificial intelligence, visualization