Abstract For many decades automated sensors, such as acoustic, optical, pneumatic, magnetic, inductive sensors and video-based image processing solutions were used for collecting data on traffic flow, speed, density, travel time etc. Using these data, the level of congestion is indirectly estimated. However, most of these sensors are costly and location-based. In addition, they cannot find the reason for congestion. To address these limitations, use of social media as a source of traffic information has been explored recently. Some of them, such as Twitter, are open source and popular throughout the world. Earlier studies that reported the use of Twitter feeds for traffic information are limited and used machine-learning techniques based on supervised learning, which needs labelled data that are difficult to obtain. Hence, this paper compares the unsupervised and supervised approaches to extract traffic-related tweets and develops a method to give the probability of level of congestion in real-time. The methods developed were applied to the tweets from Chennai city to extract the congestion information in real time, the results were validated through newspaper reports of heavy traffic congestion events. The obtained results showed a good correlation with the real congestion events.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Real-Time Traffic Congestion Information from Tweets Using Supervised and Unsupervised Machine Learning Techniques



    Erschienen in:

    Erscheinungsdatum :

    2019-10-01


    Format / Umfang :

    11 pages




    Medientyp :

    Aufsatz (Zeitschrift)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch




    Detection of Traffic Congestion from Surveillance Videos using Machine Learning Techniques

    Rao, S Govinda / RamBabu, R / Kumar, B S Anil et al. | IEEE | 2022


    Sentiment Analysis of Arabic Tweets for Road Traffic Congestion and Event Detection

    Alomari, Ebtesam / Mehmood, Rashid / Katib, Iyad | Springer Verlag | 2019


    Machine Learning Solutions to Vehicular Traffic Congestion

    Chhatpar, Pavan / Doolani, Nimesh / Shahani, Sumeet et al. | IEEE | 2018



    Real-Time Traffic Congestion Forecasting Using Prophet and Spark Streaming

    Nguyen, Ngan-Linh / Vo, Hoang-Thong / Lam, Gia-Huy et al. | Springer Verlag | 2022