Loading...

{{notif_text}}

SocialHelpouts is now CutShort! Read about it here
Who pays how much? Be informed with this salary report on Indian startups.
Why join channels?
Learn from peers
Discuss and share learning resources with the top professionals across the world
Open business or job opportunities
Earn reputation points to get consulting projects, attract talent or land jobs.
Accelerate your growth
Grow your network and get exclusive deals from our learning partners.
signup now
Manas Shrivastava asked a question

How confident are you about building AI algorithms that can predict user behaviour?`

A.I. is the new rage, companies are using it in creative ways. How likely are these algorithms going to scale?

answer
submitting answer...
submit
No answers yet. Be the first one to answer!
1 answer
Meher Vamsi ML/NLP Engineer

A.I. can scale as large as the data can scale. 

The way I think about AI is like this: 

What is the data I as a human need to predict a particular user behaviour; lets take the example of buying a product.. 
What do I need to guess:

1. Need for the product (If you need/want or not)
2. Brand, Quality, Cost etc of the product
3. Persons budget or economic situation
Given this information, i can make a pretty good guess if someone is going to buy or not.
And if a human can predict, so can a ML model given this info.

 

Now the beauty(and a little creepy) part is that we have all this data.
Let me explain:
1. The need for the product is inferred by: search queries, time you spent on looking at that product, if you added to cart or not etc.,
2. The product properties are already there in the system;
3. Economic data is also available thanks to the telecom operators, (remember the lousy form you filled when you were taking a sim card?) and things other data providers who identify you by linkedin etc.

All of this data is available with almost all the E-Commerce companies sold by 3rd party data providers.

So as long as the data required for the predictions exist, AI can scale.! 

Loading comments...
To view all answers to this question, join this channel
join this channel
Awesome! You have connected your Facebook account. Like us on Facebook to stay updated.