Using Hierarchical Time Series Clustering Algorithm and Wavelet Classifier for Biometric Voice Classification

Algorithm 1

Pseudo code of dynamic time wrap algorithm.

DTW(v₁, v₂) {

//where the vectors v₁=(a₁,…,a_n), v₂=(b₁,…,b_m) are the time series with n and m

time points

  Let a two dimensional data matrix S be the store of similarity measures

such that S[0,…,n, 0,…,m], and i, j, are loop index, cost is an integer.

  // initialize the data matrix

  S[0, 0] ∶= 0

  FOR i ∶= 1 to m DO LOOP

    S[0, i] ∶= ∞

  END

  FOR i ∶= 1 to n DO LOOP

      S[i, 0] ∶= ∞

  END

  // Using pairwise method, incrementally fill in the similarity matrix

  with the differences of the two time series

  FOR i ∶= 1 to n DO LOOP

    FOR j ∶= 1 to m DO LOOP

    // function to measure the distance between the two points

      cost ∶= d(v₁[i], v₂[j])

      S[i, j] ∶= cost + MIN(S[i-1, j],     // increment

                      S[i, j−1],   // decrement

                      S[i-1, j−1]) // match

    END

  END

  Return S[n, m]

}