Traffic offenders beware, in the near future, machine learning models will be able to predict which vehicle is more likely to commit a traffic offence.
The models will leverage large sets of data, which the intelligent machine will be able to crunch and analyse, according to researchers at the IIIT Hyderabad.
The Institute has in an analysis of traffic offences for the city of Ahmedabad, Gujarat has found that traffic offenders with more number of e-challans were less likely to pay fines. Another trend was that payment of smaller fines are preferred over larger ones.
A whopping Rs 59,24,30,400 crore is still outstanding to the government as on August 22, 2019.
To study the traffic transgression trends, the IIIT team segregated the number of paid versus unpaid challans from the total number of e-challans found in the dataset of the Ahmedabad Police department. With regarsd to the type of violations for which maximum challans were issued, the topper was jumping a red light
Challans were also characterised based on the type of offence committed – red light violation, riding without a helmet, improper parking, stop line violation, driving without a seatbelt and so on. It was found that jumping a red light accounted for the most number of challans recorded.
On pattern of offences, the study revealed a corelation to celebrations like festivals or other major events in Ahmedabad. For example, they would during Ganesh Chaturthi and Navaratri, there was a rise in challans issued. Similarly, hotspots of offences too were mapped using spatial analysis.
In addition to plotting trends, the team trained a machine learning model based on the history of registered vehicles getting challans. “Similar to how big data is being used in policing of big crimes, we have made an attempt to predict how likely a traffic offender is to commit an offence again. We achieved 95 per cent accuracy in predicting violations,” says Kanay Gupta, one of the researcher.
The researchers could narrow down on features that could predict a vehicle owner’s tendency to repeat an offence – known as ‘recidivism’. They included unpaid challans: the higher the number the greater the chances of repeat offences, number of days since last challan was issued and so on, he explained.
A major implication of the study carried out by students under Professor Ponnurangam Kumaraguru was that it can help Police get a sense of violations taking place and locations where they are happening. It can help them plan to deploy their manpower better.
The research team included--Aanshul Sadaria, Kanay Gupta, Shashank Srikanth, Hiimanshu Bhatia and Pratik Jain. They collected data from September 2015 through August 2019.
Insights from this study titled Don’t Cross That Stop Line: Characterising Traffic Violations In Metropolitan Cities” were handed over to the Ahmedabad police, who requested a similar analysis for the city of Rajkot.
With the new Motor Vehicles Act (Amendment) 2019, that came into effect from September 1, 2019 and its focus on enhancement of penalties for driving errors as well as violating other road regulations, such an analysis would be useful.