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

Algorithm 2

Preference-based task assignment algorithm (PTA).

Input: Decision sample matrix

Output: Optimal task assignment

1: Initialization

2: Normalize the sample matrix by Equations (5)–(7)

3: Determine and by Equation (8)

4: Calculate the correlation coefficient by Equation (9)

5: Build a prospective decision matrix and calculate the prospective value

6: Optimize the index weights to obtain the best comprehensive prospect value by Equations (10)–(14)

7: Calculate , , and by Equations (15)–(17), confirm the first and second value of (i.e., and )

8: for and do

9:  if only meet Condition 1 then

10:    and are compromise solutions

11:  end if

12:  else if only meet Condition 2 then

13   Calculate the largest N by , and are approximate ideal schemes

14:  end if

15:  if both meet Conditions 1 and 2 then

16:   is the optimal solution

17:  end if

18: end for