CS5691-Assignment 2 Solved

Dataset 1: 2-dimensional artificial data:  

(a)   Linearly separable data set for static pattern classification

(b)  Nonlinearly separable data set for static pattern classification

 Dataset 2: Real world data sets:                                                                                             

(a)    Image data set for static pattern classification

(b)   Image data set for varying length pattern (Set of local feature vectors representation) classification

 

Classifiers to be built for Dataset 1(a) : 

1.      K-nearest neighbours classifier, for K=1, K=7 and K=15  

2.      Naive-Bayes classifier with a Gaussian distribution for each class

a.      Covariance matrix for all the classes is the same and is2I   

b.      Covariance matrix for all the  classes is the same and is C

c.       Covariance matrix for each class is different

 

Classifiers to be built for Dataset 1(b) : 

1.      K-nearest neighbours classifier, for K=1, K=7 and K=15  

2.      Bayes classifier with a GMM for each class, using full covariance matrices

3.      Bayes classifier with a GMM for each class, using diagonal covariance matrices

4.      Bayes classifier with K-nearest neighbours method for estimation of class-conditional probability density function, for K=10 and K=20

 

Classifiers to be built for datasets (a) and (b)  in Dataset 2: 

1.      Bayes classifier with a GMM for each class, using full covariance matrices

2.      Bayes classifier with a GMM for each class, using diagonal covariance matrices

 

Use the cross-validation method to choose the best values of hyperparameters.