Decision Support Technology For City Road Traffic Congestion Safety Evacuation Based On Case Based Reasoning

With the rapid development of China's economic construction in recent years,the number of motor vehicles has been growing year by year.Although urban road infrastructure construction is constantly increasing,it is obviously unable to meet the growing demand for road traffic by motor vehicles in terms of the speed of urban road construction.In this case,traffic congestion in various cities has occurred from time to time,and has become a common problem in most cities.The social problems caused by it,such as traffic accidents,air pollution,energy loss,parking and many other security risks,have severely restricted the development of the city,and have exerted highly negative impact on economic construction as well.Thus,how to effectively relieve traffic congestion has become a top priority for transportation departments in major cities.The formation of urban road traffic congestion is multi-faceted.The congestion itself is a traffic accident.The longer the traffic congestion lasts,the greater the impact is on traffic safety.Therefore,the dredging of traffic congestion is actually to eliminate the hidden traffic safety hazards.Using case-based reasoning technology,this paper designs a decision support system for urban road guidance by investigating,collecting,sorting out and analyzing a large number of traffic congestion cases in combination with the local traffic conditions.To be more specific,this paper mainly focuses on several key issues in decision support technology of case-based reasoning,and mainly completes the following aspects:1.Collect a large number of successful solutions to ease traffic congestion and form a case base of traffic congestion counseling experts.The key factors of traffic congestion control decision-making are analyzed in detail.The TF-IDF algorithm is used to reduce the characteristic attributes of traffic congestion in the case-based decision support model.2.Study and analyze the working principle of TF-IDF algorithm,and analyze the shortcomings of TF-IDF algorithm.In the text preprocessing stage,TF-IDF algorithm optimization research is carried out to solve the weight shift problem caused by the uneven distribution of a few feature words within or between document classes according to the description of traffic congestion events.According to the optimized TF-IDF algorithm,the weight of traffic congestion feature value attributes is calculated and reduced.3.In order to improve the retrieval efficiency in the process of traffic congestion case retrieval,try to adopt the strategy of clustering first and then retrieving in case base,thereby narrowing the scope of case retrieval and improving the success rate of case retrieval.In case clustering,K-means algorithm with excellent performance in text clustering is selected to cluster traffic congestion guidance cases.At the same time,considering that the certain similarity of traffic congestion events leads to the problem of inaccurate clustering and edge cases,the K-means algorithm is optimized.The concept of micro-case cluster is introduced,the micro-case cluster of traffic congestion formed by edge cases is clustered,and then the edge cases are classified into different types of traffic congestion according to the distance of clustering center.4.The similarity calculation of case retrieval in case-based reasoning is studied in detail,and the similarity calculation method of different data are analyzed,with emphasis on distance-based similarity calculation method.5.The prototype system of urban road traffic guidance decision support is designed and developed by using case-based reasoning technology,which provides the corresponding theoretical support for effective traffic guidance,and also provides a theoretical reference for decision support technology based on text case-based reasoning.Figure[56]Table[30]Reference[141]...