A location prediction method based on automatic completion

Abstract

The invention discloses a position prediction method based on automatic completion, which comprises the following steps of S1 acquiring a spatio-temporal data set of a preset time period; S2 extracting spatio-temporal topics with common spatio-temporal characteristics; 3 selecting a training data set, training and obtaining a multi-class classifier; S4 converting the historical spatio-temporal data of the object to be predicted into spatio-temporal topic series arranged in time order; 5 predicting the future spatio-temporal theme of the current spatio-temporal theme of the object to be predicted or the spatio-temporal theme of the next few; S6 taking the position in the historical spatio-temporal data of the object to be predicted and the position of other users under the given future spatio-temporal topic as a position candidate set, obtaining the probability of each candidate position in the position candidate set, and taking the candidate position with the highest probability as thefuture position of the object to be predicted. The method of the invention considers that the human behavior mode has the characteristics of diversity and individuation, is closer to the practical application, and has better prediction accuracy and high prediction efficiency.

Description

technical field [0001] The present invention relates to location prediction technology, in particular to an automatic completion-based location prediction method suitable for predicting the next or future location of people, vehicles, equipment, events, etc. in social networks or in real life. Background technique [0002] With the development of sensing technology, there are more and more smart devices that can record the location. In the past ten years, academia and industry have collected a large amount of spatiotemporal data. From signal towers for early personal communication, GPS tracks to check-in data provided by different location-based services. Moreover, many systems originally designed for fare collection also enrich spatio-temporal data, such as smart cards in public transportation. These spatiotemporal data are known as the basis for human mobility research, including location prediction. Location prediction is considered a core function of many proactive se...

AI Technical Summary

Problems solved by technology

[0005] Although theoretical studies prove that the upper limit of predictability of human movement is 93%, the performance of current state-of-the-art position prediction algorithms is far below this upper limit

At present, there are still two major challenges to reach the theoretical upper limit of position prediction: first, how to effectively represent the movement laws embedded in spatio-temporal data; second, how to make reasonable use of these movement laws in position prediction models

Images