Research Article

The Anonymization Protection Algorithm Based on Fuzzy Clustering for the Ego of Data in the Internet of Things

Algorithm 2

Sampling time clustering algorithm.
  Input: ′, TST and ;
  Output: Substgrp; % the set of equivalent class
  BEGIN
 (1) Substgrp = ;
 (2) FOR each r in
 (3)  ;
 (4)  if then
 (5) Substgrp =
 % The set of equivalence classes with sampling interval of is
obtained. denotes the data subset whose intercept is , namely, an
equivalent class. The function of is that the number of the th equivalent class
would be counted out. denotes the th subset, namely, the
th equivalent class. It is asked that the data for each equivalence class contains at least 2 nodes.
The function of is that the number of th equivalent class. In this
paper, .
represents that the union of all subsets, namely the union of all equivalence classes. The
function is that total number of equivalence classes whose intercept is is
calculated. The maximum number of records would be covered in an equivalence class. We should
choose the set of equivalence classes which can cover the maximum number of records.
 (6)  ENDIF
 (7) ENDFOR
 END