Traffic Incident Duration Prediction using BERT Representation of Text

Agrawal, Prashansa and Franklin, Antony and Pawar, Digvijay S and Srijith, P K (2021) Traffic Incident Duration Prediction using BERT Representation of Text. In: 94th IEEE Vehicular Technology Conference, VTC 2021-Fall, 27 September 2021 through 30 September 2021, Virtual, Online.

[img] Text
IEEE_Vehicular.pdf - Published Version
Restricted to Registered users only

Download (797kB) | Request a copy


Owing to the diverse nature of traffic incidents, accepting and storing relevant data in the form of natural language is more convenient than in constrained value fields. Textual information in such cases can be rich enough for traffic incident analysis and modelling even in the absence of certain fixed set of parameters. However limited studies considered the complexity in processing such information to predict traffic incident duration. In this paper, we propose to represent the textual data from incident reports using BERT word embeddings. These text representations are then inputted into various regressors such as LSTM, XGBoost, RF and SVR to predict traffic incident duration. To demonstrate the significance of this approach, the method is compared with the state-of-the-art approach using LDA representation. Dataset used for the experiment is the Caltrans Performance Measurement System (PeMS). Result analysis indicates that the BERT- LSTM hybrid model is effective in capturing the contextual meaning of textual incident reports to predict the traffic incident duration and outperforms LDA topic modelling with MAE around 11.16 minutes. © 2021 IEEE.

[error in script]
IITH Creators:
IITH CreatorsORCiD
Franklin, Antony
Pawar, Digvijay S
Srijith, P K
Item Type: Conference or Workshop Item (Paper)
Additional Information: ACKNOWLEDGEMENT This work is a part of project “M2Smart”, supported by JST/JICA SATREPS, Japan.
Uncontrolled Keywords: BERT; LDA; LSTM; PeMS; Traffic Incidents; Word Embeddings
Subjects: Computer science
Civil Engineering
Divisions: Department of Civil Engineering
Department of Computer Science & Engineering
Depositing User: . LibTrainee 2021
Date Deposited: 06 Oct 2022 10:27
Last Modified: 06 Oct 2022 10:27
Publisher URL:
Related URLs:

Actions (login required)

View Item View Item
Statistics for RAIITH ePrint 10815 Statistics for this ePrint Item