This part requires the following WARD 1.0 dataset:
https://people.eecs.berkeley.edu/~yang/software/WAR/index.html
Unzip the databse to recognition folder, so that the directory looks like:
KCC
├───recognition
│ ├───matrix2latex
│ ├───natsortfiles
│ │ └───html
│ └───WARD1.0
│ ├───Subject1
│ ├───Subject10
│ ├───Subject11
│ ├───Subject12
│ ├───Subject13
| ...
└───tracking
└───utility
Then you may find some examples in the run_me.m file