jason balderson joseph kariampuzha joana …yahavi1/projects/dc2010t3...jason balderson joseph...
TRANSCRIPT
Jason BaldersonJason BaldersonJoseph Kariampuzha
Joana AdjaidooKofi Owusu-Boakye
Alexis Saint-Jean
Special thanks to
Ernest Chrappah
Wanda Butler
•Explore and provide insight on hidden relationships between the tickets issued, adjudication cases, and dismissed tickets.
•Data elements analyzed:
•Issuing agency,
•Violation code,
OBJECTIVE:
•Violation code,
•Violation type, adjudicated/contested, dismissed, reason for dismissal,
•Fine amount
•Etc.
•Provide MBA students an opportunity to gain applied practical experience in the field of data mining while exposing them to career options in government.
61%
39%
15%
85%
DMV Dataset Reduction
System Generated
Adjudicated
Non Actionable
ABREAST
EXPIRED METER
FAIL DISP CURR TAGS
FAIL SECURE DC TAGS
NO PARKING ANYTIME
No Standing anytime
Violation Types
Violation Types (ctd)
Adjudicators
1 2 3 4 5 6 7 8 9 10 11 12
Adjudicators (ctd)
1 2 3 4 5 6 7 8 9 10 11 12
Hearing Type
Time of Day – Photo Moving
Filter Settings
- NEW Ticket Type: (Photo Moving Violation)
Time of Day – Parking
Filter Settings
- NEW Ticket Type: (Parking Violation)
Economic Value
Economic Value (ctd)
Classification Tree - Hour
•Potential for revenue maximization by reducing the percentage of “automatic dismissals”.
•Around-the-clock enforcement of traffic and parking violations.
•Improved data-reporting systems:
SUGGESTIONS:
•Improved data-reporting systems:
•Easier to understand
•Regular evaluations
•Increased program efficiency
•Early warning system
•Training process for new adjudicators.
QUESTIONS QUESTIONS