Invited webinar -Event Detection with Temporal Predicates- by Prof. Fabio Persia

"Event Detection with Temporal Predicates" presented by Prof. Fabio Persia, DISIM, UNIVAQ; Chaired by Prof. Giovanni De Gasperis, UNIVAQ

Human perception tends to group individual values into larger structures, this is also the case for time series data. This tendency inspired us to define an event-detection language based on time intervals, which combines timepoint-based events into larger structures. Complex events can then be defined on a more abstract level by specifying temporal relationships between different time intervals. As a result, we propose a system based on an extension of relational algebra (ISEQL, Interval-based Surveillance Event Query Language), enriched with powerful temporal operators and additional constraints, which is able to: (i) extend Allen’s interval relationships; (ii) support overlap percentage and cardinality constraints over the interval relationships; (iii) provide robustness with respect to small variations in the intervals; (iv) be more efficient than state-of-the-art approaches. With the help of intuitive and interactive…