11 November 2014 3 10K Report

How can I use fuzzy clustering FCM to deal with incomplete or missing data ? I would be very grateful for answer with this simple data set, if we randomly remove amount of data from test set and use them as missing data.

i.e data set:

d=[0.3 0.4 0.3

     0.4 0.5 0.4

     0.4 0.3 0.4

     0.5 0.7 0.5

     0.5 0.5 0.5

     0.6 0.6 0.6

     0.7 0.8 0.7

     0.7 0.6 0.7

     0.8 0.7 0.8

                         ];

f=[1;1;1;2;2;2;3;3;3];

Similar questions and discussions