Estimation of Travel Time Based on Ensemble Method With Multi-Modality Perspective Urban Big Data

Discovering the Bayesian network (BN) structure from big datasets containing rich causal relationships is becoming increasingly valuable for modeling and reasoning under uncertainties in many areas with big data gathered from sensors due to high volume and fast veracity. Most of the current BN structure learning algorithms have shortcomings facing big data. First, learning a BN structure from the entire big dataset is an expensive task which often ends in failure due to memory constraints. Second, it is quite difficult to select a learner from numerous BN structure learning algorithms to consistently achieve good learning accuracy. Lastly, there is a lack of an intelligent method that merges separately learned BN structures into a well structured BN network. To address these shortcomings, we introduce a novel parallel learning approach called PEnBayes (Parallel Ensemble-based Bayesian network learning). PEnBayes starts with an adaptive data preprocessing phase that calculates the Appropriate Learning Size and intelligently divides a big dataset for fast distributed local structure learning. Then, PEnBayes learns a collection of local BN Structures in parallel using a two-layered weighted adjacent matrix-based structure ensemble method. Lastly, PEnBayes merges the local BN Structures into a global network structure using the structure ensemble method at the global layer. For the experiment, we generate big data sets by simulating sensor data from patient monitoring, transportation, and disease diagnosis domains. The Experimental results show that PEnBayes achieves a significantly improved execution performance with more consistent and stable results compared with three baseline learning algorithms.

Download Full-text

Development and Application of Dynamic Timing Optimization Platform for Big Data Intelligent Traffic Signals

E3S Web of Conferences ◽

10.1051/e3sconf/201913601008 ◽

2019 ◽

Vol 136 ◽

pp. 01008

Author(s):

Zhao Wang ◽

Mengjie Wang ◽

Wenqiang Bao

Keyword(s):

Big Data ◽

Travel Time ◽

Traffic Flow ◽

Delay Time ◽

Traffic Signals ◽

Traffic Signal ◽

Total Delay ◽

Timing Optimization ◽

Traffic Lights ◽

Signal Dynamic

As the number of car ownership increases, road traffic flow continues to increase. At the same time, traffic pressure at intersections is increasing as well. At present, most of the traffic lights in China are fixed cycle control. This timing control algorithm obviously cannot make timely adjustments according to changes in traffic flow. In this case, a large number of transportation resources would be wasted. It is very necessary to establish a dynamic timing system for Big data intelligent traffic signals. In this research, the video recognition method was used to acquire the number of vehicles at the intersection in real time, and the obtained data was processed by the optimization algorithm to make a reasonable dynamic timing of the traffic signals. The test results show that after using the big data intelligent traffic signal dynamic timing optimization control platform, in the experimental area, the overall total delay time was reduced by 23%, and the travel time was reduced by 15%. During the off-peak period, the overall total delay time in the experimental region was reduced by 17% and travel time was reduced by 10%. The big data intelligent traffic signal dynamic timing optimization platform would improve the operational efficiency and traffic supply capacity of the existing transportation infrastructure, and could provide real convenience for citizens.

Download Full-text