Stationary Classifier for Geographic Route Trace Data

Patent metadata

Field	Value
Publication number	US-2023245002-A1
Application number	US-202318298268-A
Country	US
Kind code	A1
Filing date	Apr 10, 2023
Priority date	Jun 15, 2018
Publication date	Aug 3, 2023
Grant date	—

Abstract

A first set of traces is received by a device, each trace including a set of location data. Each set of location data includes an origin point, a plurality of intermediate points, and a destination point in an ordered sequence. A bounding box surrounding a trace of the first set of traces is drawn, the bounding box corresponding to endpoints of the trace. A function determines a threshold and classifies a trace as stationary or non-stationary. A trace is classified as stationary or non-stationary by comparing a ratio represented by the trace length divided by the bounding box diagonal length to the determined threshold. In response to classifying the trace as stationary, the trace is labeled as stationary. In response to classifying the trace as non-stationary, extremities are extracted from the trace for re-classifying.

First claim

What is claimed is:

1 . A computer-implemented method for classifying activity of a device based upon data representing movement of the device, the method comprising:
receiving, by a device, a trace comprising a set of location data; drawing a bounding box surrounding the trace, the bounding box corresponding to endpoints of the trace; determining a threshold value, wherein the threshold value is determined as a function of a diagonal length of the drawn bounding box; determining whether the trace is stationary by comparing a ratio between a length of the trace and a diagonal length of the drawn bounding box to the determined threshold value; and responsive to determining the trace is stationary, labeling the trace as stationary.

2 . The method of claim 1 , further comprising:
determining the threshold value using a classification function.

3 . The method of claim 2 , further comprising:
receiving, by the device, a second trace known to be stationary; and training the classification function using the second trace.

4 . The method of claim 2 , wherein the classification function is a sigmoid function of the diagonal length of the drawn bounding box.

5 . The method of claim 2 , wherein the classification function is modified by a set of factors comprising a geolocation factor.

6 . The method of claim 1 , wherein the trace is classified as non-stationary if the ratio is greater than or equal to the determined threshold.

7 . The method of claim 1 , wherein extracting extremities comprises identifying an outlier location data point in the set of location data that is beyond a threshold radius from an intermediate location data point in the set of location data.

8 . The method of claim 1 , wherein the bounding box is drawn between two of an origin location data point, an intermediate location data point, and a destination location data point of the set of location data.

9 . The method of claim 1 , wherein the bounding box is drawn between an origin location data point and a destination location data point of the set of location data.

10 . The method of claim 1 , further comprising:
selecting an origin section for the trace, the origin section comprising an origin location data point and one or more intermediate location data points of the set of location data; removing the origin section from the set of location data; selecting a destination section for the trace, the destination section comprising a destination location data point and one or more of the intermediate location data points; removing the destination section from the trace; and dividing the trace into a plurality of subtraces, each subtrace comprising a plurality of consecutive intermediate location data points of the location data.

11 . The method of claim 10 , further comprising:
removing, from each subtrace of the plurality of subtraces, one or more intermediate location data points connecting the subtrace to one or more other subtraces of the plurality of subtraces; and storing each subtrace of the plurality of subtraces independently.

Assignees

• Mapbox Inc

Inventors

• Vollmer Ugo
• Rochon Laurier

Classifications

• G06N20/00 Primary

  Machine learning · CPC title

• H04L67/54

  Presence management, e.g. monitoring or registration for receipt of user log-on information, or the connection status of the users · CPC title

• G01S19/13

  Receivers · CPC title

• H04W4/029 Primary

  Location-based management or tracking services · CPC title

• H04W4/024

  Guidance services · CPC title