Temporal Dependency Detection Between Interval-Based Event Sequences.

We present a new approach to mine dependencies between sequences of interval-based events that link two events if they occur in a similar manner, one being often followed by the other one in the data. The proposed technique is robust to temporal variability of events and determines the most appropriate time intervals whose validity is assessed by a χ test. TEDDY algorithm, TEmporal Dependency DiscoverY, prunes the search space while certifying the discovery of all valid and significant temporal dependencies. We present a real-world case study of balance bicycles into the Bike Sharing System of Lyon.