On prediction of grouped users' trip based on massive sequence data

Abstract

Prediction of the behavior of the grouped users in the future is a meaningful research question. In this paper, we take the transit users as an example to introduce the phase space reconstruction method and use the massive sequence data to model the large-scale system with a dynamic evolution model. At the same time, considering the shortcomings of the general prediction method in large data set, an inflection point method is proposed for the automatic selection of similar points before prediction. This method does not only reduce the computational complexity of similarity in prediction process, but also significantly improve the prediction effect. Experiments show that the method proposed in this paper provides a new idea for both system modeling and group behavior predicting by using mass sequence data.