How Facebook Uses Machine Learning @ Practical Example

For Professionals and Freshers

Facebook uses machine learning in the following ways:

• Automatic friend tagging suggestions: When a pic is uploaded on facebook, a suggestion asking if you want to tag your friend in the pic appears. This is done by Facebook's face detection and recognition algorithms based on the advanced deep learning neural network research project Deepface.

• Mutual friend analysis: Facebook uses the clustering algorithm(comes under unsupervised learning) to find mutual friends.
•  
• Newsfeed: I think facebook uses ML to arrange your Newsfeed too. Like posts of close friends may come up first. Posts related to your favourite pages come up first.
•  
• Friend Suggestions: Machine learning is used by FB to suggest new friends based on mutual friend circles.

                         Practical -1 : Login to facebook  Perform following steps  

          1.Open an image in facebook and right click for Inspect Element or F12.

          2. Then check the Inspector tab. You can see the html code for that image

          3. Check the alt element content. It will give a general description 

              of the image- no of persons,   whether they/he/she are sitting,standing,laughing,

              its day or nigh   if it is a close-up pic, ..

output :

 

See html code of image carefully : specially alt content it includes description such as : 9 people ,smiling , people standing , and its night.

how facebook knows all these , answer is very simple becase of machine learning .

on the basis of past records result generated. 

                                       Practical 2: Automatic friend tagging suggestions

 

 

When i upload the photo on facebook , automatically facebook detected persons present in photo and shows there name .This is done by Facebook's face detection and recognition algorithms based on the advanced deep learning neural network research project Deepface.