BY :
RAJAT BAJAJ
101303130
COE-6
T20 PREDICTOR
T-20 PREDICTOR
 T-20 Predictor is a calculation tool
developed in WEKA , which will be
used in cricket to predict scores and
possible results of a T-20 match
depending upon certain parameters
like Venue , Pitch etc.
Objectives
1. In the first innings of a T-20 match, it acts
as a Score Predictor i.e it will predict
score at various stages for Batting Team.
2. In the second innings, it works as a Win
predictor i.e it will predict whether
Batting Team will win or lose while
chasing a Target given by the opposite
Team.
Data Selection and Collection
-> Data was
gathered from
espncricinfo.com
-> Data gathered
initially was
unstructured and
raw
-> Data was
cleaned and
duplicates were
removed
-> Only important
attributes were
taken in
consideration
-> Data was stored
Processed Dataset
Methodologies
1. Score Predictor : In this part , we will
use linear regression technique to predict
a continuous numeric value for score.
2. Win predictor : In this part , we will use
Classification using Decision Tree to
predict either win , loss or a tie.
Methodologies
 Linear Regression : It is a data mining
function that predicts a numerical value for
an attribute (to be predicted ), by using
values of other attributes
 Classification : Decision tree builds
classification models in the form of a tree
structure. It breaks down a dataset into
smaller subsets and at the same time an
associated decision tree is incrementally
developed.
Attributes
 There are total 9 attributes in the data-set
 Data-set will consist of 73 records.
 Venue has values : Home , Away and Neutral
 Pitch type has values Green , Flat , Hard and Dry
 Scores are given by their values and number of
wickets fallen
 Batting has two values : 1st and 2nd
 Result can have three values : Win , Loss or Tie
 In Phase-1 : 20-over score attribute will be class
attribute
 In Phase-2: Result attribute will be class attribute
T-20 Cricket Predicator

T-20 Cricket Predicator

  • 1.
  • 2.
    T-20 PREDICTOR  T-20Predictor is a calculation tool developed in WEKA , which will be used in cricket to predict scores and possible results of a T-20 match depending upon certain parameters like Venue , Pitch etc.
  • 3.
    Objectives 1. In thefirst innings of a T-20 match, it acts as a Score Predictor i.e it will predict score at various stages for Batting Team. 2. In the second innings, it works as a Win predictor i.e it will predict whether Batting Team will win or lose while chasing a Target given by the opposite Team.
  • 4.
    Data Selection andCollection -> Data was gathered from espncricinfo.com -> Data gathered initially was unstructured and raw -> Data was cleaned and duplicates were removed -> Only important attributes were taken in consideration -> Data was stored
  • 5.
  • 6.
    Methodologies 1. Score Predictor: In this part , we will use linear regression technique to predict a continuous numeric value for score. 2. Win predictor : In this part , we will use Classification using Decision Tree to predict either win , loss or a tie.
  • 7.
    Methodologies  Linear Regression: It is a data mining function that predicts a numerical value for an attribute (to be predicted ), by using values of other attributes  Classification : Decision tree builds classification models in the form of a tree structure. It breaks down a dataset into smaller subsets and at the same time an associated decision tree is incrementally developed.
  • 8.
    Attributes  There aretotal 9 attributes in the data-set  Data-set will consist of 73 records.  Venue has values : Home , Away and Neutral  Pitch type has values Green , Flat , Hard and Dry  Scores are given by their values and number of wickets fallen  Batting has two values : 1st and 2nd  Result can have three values : Win , Loss or Tie  In Phase-1 : 20-over score attribute will be class attribute  In Phase-2: Result attribute will be class attribute