- Patent Title: Machine learning-based geolocation and hotspot area identification
-
Application No.: US14800621Application Date: 2015-07-15
-
Publication No.: US10440503B2Publication Date: 2019-10-08
- Inventor: Pablo Tapia
- Applicant: TUPL, Inc.
- Applicant Address: US WA Bellevue
- Assignee: TUPL, Inc.
- Current Assignee: TUPL, Inc.
- Current Assignee Address: US WA Bellevue
- Main IPC: H04W24/00
- IPC: H04W24/00 ; H04W4/021 ; H04W64/00 ; H04W24/02 ; H04W24/10

Abstract:
Machine-learning based geolocation techniques may be used to provide the geolocations of user devices and determine the locations of hotspot areas. A coarse geolocation of a user device may be determined based on the wireless communication network usage information of the user device. Device data that includes the coarse geolocation of the use device may be inputted into a trained geolocation model of a machine learning algorithm. A refined geolocation of the user device that is more accurate than the coarse geolocation of the user device may be determined by using the machine learning algorithm to process the device data via the trained geolocation model. The refined geolocation of the user device may be further stored in a data store.
Public/Granted literature
- US20160021503A1 MACHINE LEARNING-BASED GEOLOCATION AND HOTSPOT AREA IDENTIFICATION Public/Granted day:2016-01-21
Information query