Our experiments rely on a combination of machine-learning (CRF) and rule-based (HeidelTime) systems. First, a CRF system identifies both EVENTS and TIMEX3, along with polarity values for EVENT and types of TIMEX. Second, the HeidelTime tool identifies DOCTIME and TIMEX3 elements, and computes DocTimeRel for each EVENT identified by the CRF. Third, another CRF system computes DocTimeRel for each...