Stay and Trajectory Information from Historical Analysis of Telecommunications Data

摘要

Sequences of location data points can be broken down into travel periods and stay periods through historical analysis of location data points. The noise and inaccuracies of location data points gathered from communications networks, such as mobile telephony networks, makes it difficult to accurately estimate when a user has stayed or dwelled at a particular location. The stay analysis may generate clusters of sequential location coordinates and may identify data points that appear to show movement but are likely to be noise, which can be artifacts of the communications network. Further, stay or travel sequences may initially be defined using thresholds of time and distance. Such thresholds may vary from one location to another and may be gathered and optimized over time.