Customer Success: Senti AI uses Machine Learning on AWS to help ABS-CBN develop its localized social media app
Posted by Senti AI
November 18, 2020
One of the most well-known media entities in the Philippines, ABS-CBN is still dedicated to always be “in the service of the Filipino.”
The media company embarked on a mission to provide Filipinos with the most relevant information in their local communities through a hyperlocal social media application. Access to relevant and trusted information has become very crucial especially at a time when fake news and disinformation are rampant.
Senti AI, a pioneering Artificial Intelligence solutions and services company, assisted ABS-CBN in ensuring that users would immediately get verified information within and around their local community.
Senti AI developed models that can determine which users can immediately access content related to their location and other surrounding areas.
A media company ‘upgrades’ citizen journalism
ABS-CBN is one of the largest and most trusted news stations in the Philippines for decades. In the 2000s, the company also introduced its citizen journalism initiative, allowing the common Filipino to play a role in sharing relevant information in their community.
ABS-CBN’s latest project, SHAREA, comes in the form of a social media application, which is envisioned to build stronger location-based communities.
The social media application, however, needed a technology that can easily distribute news and user-generated content to relevant audiences. To address this gap, ABS-CBN partnered with Senti AI to create a system that personalizes which information users can see first, depending on their location.
Senti AI’s models do more with less through AWS
To enable the distribution of content based on the closeness of the location to the original poster, Senti AI developed various Machine Learning models that can match users with relevant content based on hyperlocal parameters.
The models Senti AI helped develop used the capabilities of Natural Language Processing and Machine Learning to understand localized content.
To make sure that the models are built to scale and support high-volume users, Senti AI’s models were deployed in production using Kubeflow on Amazon EKS. This choice of technology allows ABS-CBN to cut operation costs and manually turn on and off their Machine Learning tools
As of early November, ABS-CBN’s SHAREA social media application has gone live already. In its early access phase, the app saw more than 10,000 downloads in the Google play store.
Want to know more?
Get more details on what Voix is and how it can help your organization achieve better customer experience