Application

Event recognition for public space surveillance using the CAVIAR benchmark dataset.

Artikis A., Sergot M. and Paliouras G. An Event Calculus for Event Recognition. IEEE Transactions on Knowledge and Data Engineering (TKDE), 27(4):895-908, 2015.

Directory Structure

• /CE patterns. Original and compiled CE patterns along with declarations.
• /data. The complete CAVIAR dataset in RTEC format.
• /execution scripts.

Execution Instructions

Unzip the compressed dataset in the /data/complete caviar/ directory.

Go to the /execution scripts/complete caviar/ directory.

Launch Prolog.

Load continuousQueries.prolog:

['continuousQueries.prolog'].

Invoke continuousER(TimesFile, InputFile, Window, Step, LastTime) with appropriate values for the 4 arguments. Eg:

continuousER('times.txt', 'input.txt',10000,10000,1007000).

instructs RTEC to perform continuous queries where windowstep10000, the event recognition times per query are recorded in times.txt, the number of input events per window are recorded in input.txt, and querying ends when time-point 1007000 is reached.

To see the intervals of the fluents in the last window, type:

holdsFor(U,I).