A Random Forest Model for Travel Mode Identification Based on Mobile Phone Signaling Data

Lu, Zhenbo; Zhang, Long; Xia, Jingxin; An, Chengchuan

doi:10.3390/su11215950

Cited by 27 publications

(14 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The proposed system applied to the Kunshan data set in China and the model produces better accuracy 90%. This accuracy level is suitable for all kinds of transport modes except for buses [9].…”

Section: Literature Reviewmentioning

confidence: 99%

Traffic Prediction System Using Machine Learning Algorithms

Ramchandra¹,

Rajabhushanam²

2021

Proceedings of the First International Conference on Computing, Communication and Control System, I3CAC 2021, 7-8 June 2021, Bh

View full text Add to dashboard Cite

Traffic congestion is defined as the state on transport which is characterized by slower speeds of vehicles this is also because of the bad condition of the roads, weather, concern zone, temperature, etc. This traffic flow prediction is mainly based on the realtime dataset which is collected with the help of various cameras and sensors. In recent day the deep learning concepts has dragged the attention for the detection of traffic flow predictions. In this paper, some of the common and familiar machine learning concepts like Deep Autoencoder (DAN), Deep Belief Network (DBN), and Random Forest (RF) are applied on the online dataset for the traffic flow predictions. The important attributes of weather, temperature, zone name, and day are used to predict the traffic flow of the particular zone. The performance of the proposed system can be evaluated by using accuracy, precision, and RMSE, and MSE value. Among the three methods, the DT technique produces a better result.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Traffic Prediction System Using Machine Learning Algorithms

Ramchandra¹,

Rajabhushanam²

2021

Proceedings of the First International Conference on Computing, Communication and Control System, I3CAC 2021, 7-8 June 2021, Bh

View full text Add to dashboard Cite

show abstract

“…However, the focus of this study is mainly on extracting the complete travel trajectory of subway passengers when the mobile phone data of subway passengers have already been distinguished from all the data collected, rather than focusing on the travel mode identification. In future studies, one can also use some other advanced and sophisticated models to identify different travel modes with mobile phone data [33][34][35].…”

Section: Case Studymentioning

confidence: 99%

Extracting the Complete Travel Trajectory of Subway Passengers Based on Mobile Phone Data

Zhang

Cheng

et al. 2022

Journal of Advanced Transportation

View full text Add to dashboard Cite

The usage of mobile phones has undergone tremendous growth in the past decades. Large amounts of mobile-phone signaling data (MSD) are generated while using various mobile phone applications. The large-scale MSD presents opportunities for transport planners to utilize it for better planning and management of the transportation system. In this paper, we use MSD to analyze subway passengers’ travel behavior and extract their complete travel trajectories. The complete travel trajectories of subway passengers include their trajectories, both inside and outside the subway system. In the first stage, the MSD from the subway base stations is selected, sorted by time, and the rough trajectory in the subway system is extracted. The ground base stations around the subway station are then considered to correct the boarding and alighting subway stations in order to obtain a more detailed trajectory. In the second stage, the service range of the base station is determined according to the Thiessen polygon, and a temporal dynamic threshold is proposed to extract the passenger’s stop point outside the subway system. Finally, the complete trajectories of subway passengers are obtained. The proposed algorithms are verified using a set of MSD collected in Suzhou, China. The results show that the proposed algorithms can effectively extract the complete travel trajectory of subway passengers.

show abstract

“…Lu et al, Cheng et al, and Semanjski et al applied an RF algorithm to identify the travel mode and analyze travel choice behavior. Compared with other machine-learning algorithms such as support vector machine (SVM), the results indicate that the RF method has a stronger capability to deal with multi-dimensional data or mixed types of data, such as cellular phone data, built environment data, and travel survey data (19)(20)(21). Zhang and Haghani used this method to improve the highway travel time prediction model (22).…”

Section: Literature Reviewmentioning

confidence: 99%

“…The relative importance of four features can be evaluated by the RF model. A higher value of relative importance indicates stronger influences on the model's prediction results (19). The importance evaluation result is shown in Table 3.…”

Section: Determination Of Model Parametersmentioning

confidence: 99%

Random Forest Model for Trip End Identification Using Cellular Phone and Points of Interest Data

Yang

Wang

Jin

et al. 2021

Transportation Research Record: Journal of the Transportation R

View full text Add to dashboard Cite

Cellular phone data has been proven to be valuable in the analysis of residents’ travel patterns. Existing studies mostly identify the trip ends through rule-based or clustering algorithms. These methods largely depend on subjective experience and users’ communication behaviors. Moreover, limited by privacy policy, the accuracy of these methods is difficult to assess. In this paper, points of interest data is applied to supplement cellular phone data’s missing information generated by users’ behaviors. Specifically, a random forest model for trip end identification is proposed using multi-dimensional attributes. A field data acquisition test is designed and conducted with communication operators to implement synchronized cellular phone data and real trip information collection. The proposed identification approach is empirically evaluated with real trip information. Results show that the overall trip end detection precision and recall reach 95.2% and 88.7% with an average distance error of 269 m, and the time errors of the trip ends are less than 10 min. Compared with the rule-based approach, clustering algorithm, naive Bayes method, and support vector machine, the proposed method has better performance in accuracy and consistency.

show abstract

A Random Forest Model for Travel Mode Identification Based on Mobile Phone Signaling Data

Cited by 27 publications

References 35 publications

Traffic Prediction System Using Machine Learning Algorithms

Traffic Prediction System Using Machine Learning Algorithms

Extracting the Complete Travel Trajectory of Subway Passengers Based on Mobile Phone Data

Random Forest Model for Trip End Identification Using Cellular Phone and Points of Interest Data

Contact Info

Product

Resources

About