Algorithm 1: Computing weight vector for EROS (Yang and Shahabi, 2004)

Input: An n × N matrix S, where n is the number of variables for the dataset and N is the number of MTS items in the dataset.

Each column vector si in S represents all the eigenvalues for ith MTS item in the dataset. sij is a value at column i and row j in S. s*i is ith row in S. si* is ith column

1: for i = 1 to N do

2:       $s_i\leftarrow s_i/\sum _{j=1}^n{s}_{ij}$

3: end for

4: for i = 1 to n do

5:       wi ← f(s*i)

6: end for

7: for i = 1 to n do

8:       $w_i\leftarrow w_i/\sum _{j=1}^n{w}_j$

9: end for