South African student Taahir Patel is tapping into Twitter data to map hijacking hotspots across the country.
SA is experiencing a steep increase in the number of hijacking incidents, recording 5 866 cases between April and June – a 14% increase from the previous corresponding period.
To assist in reducing the high number of hijacking incidents, Patel, a computer science student at the University of the Western Cape, is using micro-blogging platform Twitter to plot the occurrence of hijackings on a map. This data can then be used to inform other users, notify emergency responders and help law enforcement create incident reports.
His work is now being turned into a mobile application for general public use, to inform them of the dangers on the road.
Patel is Telkom’s guest at the annual Southern Africa Telecommunication Networks and Applications Conference (SATNAC), hosted by the telephony group, where he is showcasing his innovative solution.
Telkom is hosting Patel and 44 other computer science and engineering graduates who are presenting at the Telkom Centres of Excellence programme held at the conference.
Launched in 1997, the initiative provides a platform for future industry leaders to present their research to current industry giants.
“The risk of hijackings is becoming greater,” notes Patel. “But what if there was a way we could mitigate this risk using a platform that people already trust?”
The enterprising student says he tested a combination of three machine learning techniques – Multilayer Feed-forward Neural Network (MLFNN), Convolutional Neural Network, and Bidirectional Encoder Representations from Transformers – to separate the topic of “hijacking” from actual, relevant incident reports on Twitter.
“The trick is knowing what Twitter data is relevant and what simply a tweet is. Aside from character limit, Twitter doesn’t have many restrictions – so it’s crucial that we put the right parameters in place to extract valid data.
"The MLFNN technique achieved 98.99% accuracy in determining the validity of tweets as hijacking reports. This is an extremely encouraging result and proves that we can successfully use social media data to develop a map that indicates hijacking hotspots.”
Share