A Task Assignment Method Based on User-Union Clustering and Individual Preferences in Mobile Crowdsensing

Algorithm 1

User-union clustering algorithm (UCA).

Input: User data , task set , maximum user limit τ

Output: Set of K task clusters

1: ProperCluster(x_i)

2: Determine the center of the initial task sets and user data evaluation indexes (

3: Calculate the user similarity by Equation (1), and sort data in descending order

4:  for DO

5:   if

6:      enter

7:      and are saved in

8:     break

9:   else

10:     if is less than the minimum similarity value in

11:      continue

12:     else

13:       joins the jth union and deletes edge user ()

14:      ProperCluster()

15: repeat

16:  for DO

17:   ProperCluster(x_i)

18: until saturate task requirements or reach the maximum number of iterations

19: End