Data Science on Facebook :
Facebook has been the social media leader of the world as to date. People are spending a lot of time on social media uploading and posting their personal stories like relationships, parties, weddings, and more information about themselves.
For instance, Kenya has 7 million Facebook users as per research on Africa 2020 population and internet users –Facebook subscribers as at 31st December 2019.
With these millions of users in Kenya and the rest of the world, Facebook utilize a large scale of quantitative research and analysis through Data science to gain insights about the social interaction of the people across the world. It has used advanced data science techniques like deep learning used for fiscal recognition text analysis among others. Some of the real-life examples that show how Facebook uses data Science are the flashbacks on celebrating anniversaries by making slideshows of photos and posts that received the most number of likes and comments, suggesting to you the people you may know and people around you if your location in the device one is using is turned on, notifications on how long you’ve been friends to someone on Facebook, memories
Facebook uses Deep Face which is a technology they have developed that has facial recognition and image processing capability. Facebook can track its users across the internet and other posts with image data provided by users through the sharing of their image posts.
It is said that this technology tool is the most advanced and powerful for face recognition. it is more successful than humans in recognizing two or more images of the same person.
This tool is said to have gone too far as it would allow Facebook, based on a high-resolution photograph of a crowd to put names to many of the faces captured on the photo.
Large shared data on Facebook is in text form. Facebook uses a tool named Deep Text. It is a tool developed to extract meaning from words we post by analyzing them contextually. The tool has the ability to explain the meaning of every word based on how it has been used meaning; it isn’t tripped up by variations in spelling, dialects and the language used. Facebook said that “language is agnostic” due to the way it assigns labels and can easily switch between working across different human languages. The tool can is also been used to direct people toward products that they may want to purchase based on conversations they are having.
They also use a neural network to analyze the relationship between words to understand their meaning.
Facebook is more of an advertising corporation, therefore it uses deep neural networks to decide which adverts to show to which users depending or the conversations, likes, and posts its users do. This has been a cornerstone of many businesses been advertised on the Facebook platform.