Abstract:
How to extract great insights from satellite imagery? Last few years have seen enormous growth on AI-based data-enrichment techniques. Automatic sentiment and topic identification in text documents have become ubiquitous. Image tagging and brand-detection are now mainstream. Taking advantage of these techniques, we trained a model capable of identifying the Socioeconomic Status of a delimited region relying solely on satellite imagery. The model, when joined with other geographical features, can be used to create rich attractiveness models, sampling frameworks and more. New advances in Artificial Intelligence through Machine Learning have pushed forward Computer Vision. In this section we will describe how the great availability of image-data and new computer algorithms have triggered a revolution on Computer Vision. We will also explore how this new models can be leveraged in a market-research context.
This could also be of interest:
Videos
Machine learning evolving computer vision (Spanish)
Catalogue: Latin America 2018: Market Intelligence Festival for Latin America
Authors: Mario Abarca, Gustavo RamÃrez
Company: Kantar
May 14, 2018
Research Papers
The value of machine learning in privacy
Catalogue: Fusion 2018 (Big Data World + Global Qualitative)
Authors: Carlos Ochoa, Ezequiel Paura
Company: Netquest
November 11, 2018
Research Papers
Can AI (Computer Vision) Help Us to Evaluate Creatives at Large Scale?
Catalogue: Congress 2023
Author: Minh Nguyen
 
September 27, 2023
![The documents and videos are available for ESOMAR members only!](/static/img/ana_document_bg.png)