Method for predicting crowd aggregation trend based on ARMA algorithm

Abstract

The invention discloses a method for predicting a crowd aggregation trend based on an ARMA algorithm. In order to search for a more effective implementation scheme for crowd aggregation trend prediction, the method comprises the following steps: obtaining current day activity information of a to-be-predicted day of a preset place and daily aggregated person values and daily activity information within p days before the to-be-predicted day; obtaining a predicted aggregated person value of the to-be-predicted day based on the current day activity information, the daily aggregated person value within p days and the daily activity information; obtaining a daily aggregated person value in r days before the to-be-predicted day, determining whether the predicted aggregated person value is abnormal or not based on the daily aggregated person value in r days and the predicted aggregated person value, and if yes, displaying that the predicted aggregated person value is abnormal. According to themethod, the long-time multi-dimensional attention crowd place aggregation data is adopted, the ARMA algorithm is used for extracting the potential correlation between the event which cannot be judgedmanually and the time, and the defect of experience judgment on the attention crowd aggregation degree in traditional work is overcome.

Description

technical field [0001] The present invention relates to the field of big data, in particular, the present invention relates to a method for predicting crowd gathering tendency based on ARMA algorithm. Background technique [0002] When carrying out actions such as major event predictions and security in public places, the gathering of people of concern can often reflect the behavior of the objects of concern, and at the same time, the degree of gathering of people of concern often has a certain correlation with the occurrence of events in the target area. To this end, it is necessary to be able to conduct focused research on the aggregation trend of the population of concern. [0003] The traditional artificial experience-based research on the gathering of people of concern can achieve results to a certain extent, but there are two obvious problems at the same time: the first is the ability to predict group events in advance. Aggregation often depends on the researchers' ow...

AI Technical Summary

Problems solved by technology

[0003] The traditional artificial experience-based research on the gathering of people of concern can achieve results to a certain extent, but there are two obvious problems at the same time: the first is the ability to predict group events in advance. Aggregation often depends on the researchers' own experience, focusing on the management and control of places of concern to group events and crowded places, but this method cannot effectively predict events outside of experience; the second is the setting of the aggregation threshold , in traditional work, the gathering threshold for crowds of concern is manually set, and the criteria for judging gatherings may change over time and with changes in places, so at this point artificial settings cannot well divide normal gatherings and abnormal mass event aggregation

Images