| Whether the first-aid vehicle runs smoothly in the pre-hospital first-aid stage determines whether the patient can be treated in the golden time of rescue.Therefore,in view of the increasingly severe road congestion problem,the decision basis is provided for the first-aid vehicle to choose the road to avoid the congested road to some extent.Therefore,it is of great practical significance to make full use of the GPS big data of xi ’an 120 center for data mining and research on traffic condition prediction and classification.Firstly,the data structure was analyzed according to the characteristics of the emergency vehicle GPS data,the existing problems were detected and repaired,and the traffic flow parameters were extracted and the traffic status was graded.Secondly,on the study of traffic forecast and classification,on the basis of choosing appropriate parameters of traffic flow prediction model and traffic classifier——kalman filtering model and the naive bayes classifier,and carries on the summary and analysis,thus put forward in this paper,for the improvement of traffic classifier direction: according to different attribute variables,according to the classification of the importance of giving different weights.On the basis of obtaining Fisher’s weight,the genetic simulated annealing algorithm is used to find the optimal weight.Based on the theory of traffic condition prediction and classification,the extracted traffic flow parameters were analyzed experimentally,the parameters of kalman filter model were adjusted according to the characteristics of the data,and the traffic flow parameters at time t(10)1 were predicted.By dividing different data sets,the classification accuracy of the improved naive bayesian classifier is verified to be improved by 2.56%,and the optimal prediction results of the kalman filter model are also verified.Finally,the prediction and classification of pre-hospital emergency traffic condition was completed. |
PDF Full Text Request