Real-Time Forecast of Tourists Distribution Based on the Improved k-Means Method

Algorithm 1

K-means clustering: KM(𝑂,𝑘).

Input: the dataset O, and the number of clusters k

Output: Set of k clusters 𝑆={𝑆₁,𝑆₂,…,𝑆_𝐾}

Initialization: 𝑠={}, Randomly, select k initial centroids

Begin

 Repeat

   Step  1. Vectors are assigned to the closest centroid

   Step  2. The objective function J is computed

   Step  3. New centroids are calculated as the mean of the vectors assigned to each cluster

  Until Convergence (or no change in the objective function)

  Return S

  End