Geo-analytics elucidated

Updated: Feb 27, 2020

Geo-analytics is the intersection of GIS (Geographic Information Systems) and data analytics. The GI (Geographic Information) Sciences is an interdisciplinary of specific methods in geo-statistics, spatial analysis, remote sensing, data engineering, network analysis etc. It is specifically the study of methods for best modeling data in a GIS.

A GIS is an integration with IT technology. It’s a way of architecting the business systems to cooperate with spatial information i.e. using a relational database for querying, modeling and visualizing insights decrypted from big data.

Here are some common examples to cite for the usage of geo-analytics as data analysis, data visualization and data modeling tool:

mapping election data, tracking the spread of a virus, understanding the cause of earthquakes and tracking bird migration etc.

One way to categorize data problems is by understanding how the data can be expressed for a physical feature:

[1] Tangible: A tangible network problem deals with understanding traffic congestion at certain points in a road network (maybe to ultimately resolve traffic jams.)

[2] Intangible: An intangible network problem deals with the weighted influence that pairs of (origin-destination) have, on total delay in a road network (maybe to elucidate where the major influences are and whether or not those routes have an anomaly, or if investment into those routes may improve the network.)

While both problems examine data captured from the performance of a physical road network, the Tangible example describes static observations and connects those data points to a physical road structure at a certain place and time.

The Intangible problem deals with assigning the amount of influence that one trip (pairs of origin-destination) has on other trips, and ultimately how it impacts the entire road network. In this example, each edge (polyline) of the road network holds a weight that is determined by how much it influences another edge in the road network.

This is an intangible weight because the number to be used in linear regression equations is based on an interaction between two points rather than a stationary observation at one point or the other. So an intangible weight is assigned to it that represents its degree of influence on the entire network.

Interested to discuss any related business problem. Write to us at

99 views0 comments

Recent Posts

See All