QUORUM Perilog encompasses four key data mining techniques: keyword-in-context search, phrase search, phrase generation, and phrase discovery. Together, these methods facilitate in-depth analysis of extensive collections of aerospace incident narratives. QUORUM Perilog, often referred to simply as Perilog, takes advantage of the situational structure found in “unstructured” narrative incident reports. By modeling the contextual frameworks of these narratives, it also captures the underlying structures of the incidents themselves. This capability allows for effective exploration of narrative databases. Developed by NASA, Perilog was initially created to assist the FAA’s Aviation Safety Reporting System (ASRS).
