No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Urban Traffic Congestion Discrimination Algorithm Based on the Ordered Decision Theory
Abstract
Congestion discrimination is the basis to effectively develop traffic control strategies. This paper takes data from sensors as the research object, based on ordered decision theory to sort the urban network traffic congestion indicators and predict traffic congestion situation based on decision tree algorithm, consequently, indicator set which can describe the performance of network links and intersections is obtained. The proposed method reveals that there is an ordered relation between indicators and traffic congestion. By eliminating redundant indicators, this algorithm can get the closely related indicator subset to determine whether traffic congestion happen or not, accordingly the effectiveness of traffic congestion discrimination is improved.
ResearchGate has not been able to resolve any citations for this publication.
In many classification problems the domains of the attributes and the classes are linearly ordered. For such problems the classification rule often needs to be order-preserving or monotonic as we call it. Since the known decision tree methods generate non-monotonic trees, these methods are not suitable for monotonic classification problems. We provide an order-preserving tree-generation algorithm for multi-attribute classification problems with $k$ linearly ordered classes, and an algorithm for repairing non-monotonic decision trees. The performance of these algorithms is tested on a real-world financial dataset and on random monotonic datasets.
For classification problems with ordinal attributes very often the class attribute should increase with each or some of the explaining attributes. These are called classification problems with monotonicity constraints. Classical decision tree algorithms such as CART or C4.5 generally do not produce monotone trees, even if the dataset is completely monotone. This paper surveys the methods that have so far been proposed for generating decision trees that satisfy monotonicity constraints. A distinction is made between methods that work only for monotone datasets and methods that work for monotone and non-monotone datasets alike.
In order to quickly identify traffic congestion from mass dynamic traffic information) traffic congestion pattern and the characteristics of various data mining technologies were analyzed, an auto-identifying method of urban expressway traffic congestion was designed. The flow, speed and occupancy of expressway were combined into several new eigenvectors, optimized multi-layer feedforward perceptron model was adopted to classify the eigenvectors during congestion and non-congestion, recurrent congestion and non-recurrent congestion could be distinguished by analyzing the variances of the model outputs, the method was tested with simulated data and actual data from an urban expressway. The result shows that the method has great practicability and can identify congestion states on urban expressway correctly.
Authors studied the basic relationship between flow and occupancy to propose a new criterion for identifying the onset of congestion in the road networks. With the criterion, authors derived some new guidelines for detecting congestion incidents, and could also measure off the traffic stream into four stages. Finally, with the research in above, authors advanced a new detection algorithm for road traffic congestion incidents which was validated by using the data from actual investigation.
Multi-agent based simulation has been looked as an efficient tool for large-scale system, like urban intelligent transport system. However, the main problem is how to build a multi-agent based model to simulating it. This research presents a multi-agent based demand responsive transport (DRT) services model, which adopts a multi-agents planning approach for metropolis traffic services control. In this paper, we propose an agent based multi-layer distributed hybrid planning model for the real-time problem. In the proposed model, there are three layers: transport-admin agent layer, node-station agent layer and taxi agent layer. The agent for station and each vehicle have a planning domain, and each agent for vehicle to find a set of routes by its local search, and select a route by a cooperation of agents in its planning domain. Based on this model, a simplified multi-agent based demand responsive transportation services system can be developed that is effective for reducing traffic congestion and air pollution. By computational experiments, we examine an effectiveness of the proposed method.
Traffic congestion may occur in the whole road network or in local of that network, moreover, it has big possibility to arouse second congestion. Based on the Davison modified model of BRP function and the definition of road service standard, this paper analyzed the runtime of road network. Under complexity road network, the optimal path which is chosen in normal traffic environment maybe not the optimal solution in traffic congestion environment because of the differences of road service level and dynamic development of traffic flows in the crossroads. Considerate the reliability of the chosen path, this article built a static traffic distribution model to analysis urban transportation with macroscopic perspective, and provide suggestions for relieve urban traffic congestion.
Ordinal classification plays an important role in various decision making tasks. However, little attention is paid to this type of learning tasks compared with general classification learning. Shannon information entropy and the derived measure of mutual information play a fundamental role in a number of learning algorithms including feature evaluation, selection and decision tree construction. These measures are not applicable to ordinal classification for they cannot characterize the consistency of monotonicity in ordinal classification. In this paper, we generalize Shannon’s entropy to crisp ordinal classification and fuzzy ordinal classification, and show the information measures of ranking mutual information and fuzzy ranking mutual information. We discuss the properties of these measures and show that the proposed ranking mutual information and fuzzy ranking mutual information are the indexes of consistency of monotonicity in ordinal classification. In addition, the proposed indexes are used to evaluate the monotonicity degree between features and decision in the context of ordinal classification.