Interesting Activities Discovery for Moving Objects Based on Collaborative Filtering

Algorithm 1

Interesting activity discovery (IAD).

Input: 𝑂_id, k, ID_th

Output: the latent activity set of 𝑂_id

01: Abstract all the activity sequences and stored them in the 𝑆_set;

02: Build object-visit-matrix M according to the 𝑆_set;

03: Compute object-region interest degree from M, and build the interest degree matrix 𝑀_id;

04: Search similar objects using hybrid CF algorithm, and generate similar matrix 𝑀_𝑠 according to 𝑀_id;

05: Find the k most nearest neighbors to 𝑂_id from 𝑀_𝑠;

06: On the basis of the k most nearest neighbors, recommend the latent activities for 𝑂_id;

07: If the 𝑂_id’s interesting degree to the recommended activities is less than ID_th, then remove the recommended activity,

   otherwise, recommend them to 𝑂_id.

End.