Data processing method and device for real-time prediction of number of airport security inspection people

Abstract

The invention discloses a data processing method for real-time prediction of the number of airport security inspection people. The method comprises the steps that preprocessing, including data cleaning, structured processing and data integration, is performed on data in use; feature extraction is performed on the preprocessed data; extracted features are input into multiple prediction sub-models for prediction of the number of the airport security inspection people, and prediction results are output respectively; the prediction results output by the prediction sub-models are evaluated respectively, weight values are given to all the prediction sub-models according to the evaluation results, and a combination model is formed; and prediction of the number of the airport security inspection people is performed according to the combination model, and a matching prediction result is output. According to the method, prediction is performed by the adoption of multi-model combination, prediction results of a time sequence model, a random forest and a GBDT are used for evaluation, then assignment is performed on the weight of each model, and the combination prediction model is formed, so that higher prediction accuracy is achieved. Meanwhile, the invention furthermore provides a processing device for real-time prediction of the number of the airport security inspection people.

Description

technical field [0001] The invention relates to the field of data processing, in particular to a data processing method and device for real-time prediction of the number of people in airport security checks. Background technique [0002] The airport has a huge passenger throughput. Airport services such as security check, security, and emergency response all hope to predict the future passenger throughput and deploy manpower and material resources in advance to better serve passengers and respond to emergencies. . [0003] The International Air Transport Association (IATA) predicts that the average annual growth rate of passenger traffic demand will reach 3.7% in the future, and the demand for air travel will double in the next 20 years. At present, the quality of my country's air transport services is not enough to match the status of a major civil aviation country. Compared with developed civil aviation countries, there is a certain gap in service level and quality, and i...

AI Technical Summary

Problems solved by technology

[0006] The technical solutions disclosed in the above two patents and the existing technical solutions for predicting airport passenger flow have the following defects: first, there are certain limitations in the use of data; The data used in the prediction research either come from the real-time data of the check-in counter or the flight plan data, which are not enough to make accurate predictions for the security personnel; thirdly, most airports currently base the security checkpoints on the basis of the planned departure volume of flights. Forecasting the number of passengers, this method cannot achieve accurate and real-time grasp of passenger flow, and it is difficult to realize the dynamic scheduling of airport security check resources; Fourth, existing airports use real-time data from check-in counters to predict the number of check-in people, but there is no such method. Considering the impact of historical data, the prediction accuracy of security check numbers cannot be guaranteed

[0007] In view of the above problems in related technologies, there is still a lack of a method of combining multiple machine learning algorithms to process the real data of airport passengers, and accurately predict the number of people in airport security checks by building a model, that is, there is a lack of data for real-time prediction of the number of people in airport security checks Treatment methods and devices

Images