The following queries can be used to see the differences (if any) that the validator and your system generated:
Accident alerts that the validator generated but yours did not
SELECT * FROM accAlertNotInOriginal;
Accident alerts that the your system generated but the validator did not
SELECT * FROM accAlertNotInValidator;
Toll alerts that the validator generated but yours did not
SELECT * FROM tollAlertNotInOriginal;
Toll alerts that the your system generated but the validator did not
SELECT * FROM tollAlertNotInValidator;
Account balance that differ from the validator and yours
SELECT * FROM type2wronganswers;
Daily expenditure that differ from the validator and yours
SELECT * FROM type3wronganswers;
If all queries return (0 rows), the two versions are 100% matched.
The following queries can be used to see data imported from your system:
Number of car datapoints
SELECT count (*) FROM input;
Car datapoints
SELECT * FROM input;
Number of accident alerts
SELECT count (*) FROM accidentAlerts;
Accident alerts
SELECT * FROM accidentAlerts;
Number of toll alerts
SELECT count (*) FROM tollAlerts;
Toll alerts
SELECT * FROM tollAlerts;
Number of account balance requests
SELECT count (*) FROM type2requests;
Account balance requests
SELECT * FROM type2requests;
Number of daily expenditure requests
SELECT count (*) FROM type3requests;
Daily expenditure requests
SELECT * FROM type3requests;
Number of account balance answers
SELECT count (*) FROM outputtype2;
Account balance answers
SELECT * FROM outputtype2;
Number of daily expenditure answers
SELECT count (*) FROM outputtype3;
Daily expenditure answers
SELECT * FROM outputtype3;
The following queries can be used to see data the validator generated:
Number of accident alerts
SELECT count (*)
FROM tollAccAlerts
WHERE accidentSeg <> -1; If the two versions are 100% matched, this number must be equal to the the number at (2.3)
Accident alerts
SELECT time, carid, accidentSeg
FROM tollAccAlerts
WHERE accidentSeg <> -1;
Number of toll alerts
SELECT count (*) FROM tollAccAlerts; If the two versions are 100% matched, this number must be equal to the the number at (2.5)
Toll alerts
SELECT time, carid, lav, toll
FROM tollAccAlerts;
Number of account balance answers
SELECT count (*) FROM type2answer; If the two versions are 100% matched, this number must be equal to the the number at (2.11)
Account balance answer
SELECT * FROM type2answer;
Number of daily expenditure answers
SELECT count (*) FROM type3answer; If the two versions are 100% matched, this number must be equal to the the number at (2.13)
Daily expenditure answers
SELECT * FROM type3answer;
The following queries can be used to see other information that the validator extracted
Accidents
SELECT * FROM accident ORDER BY firstminute;
Statistics
SELECT * FROM statistics;
Please address all questions about the web site and
the benchmark tools to Nga Tran at nga@cs.brandeis.edu.