Times are displayed in (UTC-05:00) Eastern Time (US & Canada)Change
A Spotters Guide to Errors and Fraud in Human Movement Data
Topics: Digital Geographies
, Geographic Information Science and Systems
, Marketing Geography
Keywords: Human Movement, Cell Phone, Big Data, Machine Learning Session Type: Virtual Paper Abstract Day: Tuesday Session Start / End Time: 3/1/2022 05:20 PM (Eastern Time (US & Canada)) - 3/1/2022 06:40 PM (Eastern Time (US & Canada)) Room: Virtual 57
Authors:
Peter E Lenz, Near Pte Ltd
,
,
,
,
,
,
,
,
,
Abstract
Electronic device derived location information has provided human geographers a scale and speed of data previously the domain of those involved in the physical sciences.However, because HMD is sourced from a highly distributed network of unreliable sensors (i.e. cell phones) the data quality is highly variable. Outright hostile actors also operate in this space and can impact this supply of data. This talk will review several kinds or erroneous, fraudulent, or just dang weird patterns that have been observed in trillions of individual events collected by the data science team at Near, a decade-old global aggregator of this data, in a effort to disseminate knowledge of these phenomena to fellow geographers as Human Movement data becomes more heavily used for analysis.
A Spotters Guide to Errors and Fraud in Human Movement Data
