Loading...

{{notif_text}}

Find jobs
Hire talent
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
{{::getFormatedLocalTime("2017-04-21T11:20:23.886Z", {without_time: true})}}

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
cancel
No answers yet. Be the first one to answer!
1 answer
Meher Vamsi ML/NLP Engineer
{{::getFormatedLocalTime("2017-05-29T14:06:50.351Z", {without_time: true})}}

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.! 

Comment
Load more comments
Loading comments...
To view all answers to this question, join this channel
join this channel
Learn how Cutshort Channels work
Other questions
Siddhant askedon {{::getFormatedLocalTime("2020-03-17T15:33:33.217Z", {without_time: true})}}
How do I find jobs in data science in Bangalore?
Where can i find Data science jobs at startups in bangalore?
0
1
Saurabh askedon {{::getFormatedLocalTime("2018-10-22T12:29:11.677Z", {without_time: true})}}
Any openings in data science and machine learning? LinkedIn: https://www.linkedin.com/in/saurabh-kamble-52791314a/
I have 12 months of Experience with Python[Numpy,Pandas,MatplotlibSeaborn,Sklearn] and R programming. I have worked on Machine Learning , Natural Language Processing, Deep learning, Data Visualization tools like Tableau , QlikSense , Advance Excel and MySQL, looking for applying these skills on real-time data.Technical Skills:Applied Statistical Modelling for models analysing Z-score, Skew and Kurtosis, Data Transformation, Hypothesis Testing, Using ML Algorithms like Linear/Logistic Regression, Decision Trees, Random Forest,Xgboost,Naive-Bayes,PCA,SVM,K-means,KNN including working on  Natural Language Processing with TensorFlow. Sentiment Analysis on IMDB movie dataset, Image Classification using CNN, Customer Segmentation/Churn Analysis.       
0
0
Anubhav askedon {{::getFormatedLocalTime("2018-06-20T05:06:26.239Z", {without_time: true})}}
Why companies prefer experience in Python over R in a data science interview?
Nearly 80% of the respondents in our ongoing survey say that recruiters prefer Python over R in a Data science interview. What's been your experience?
3
4
Nikunj askedon {{::getFormatedLocalTime("2018-05-29T14:06:20.811Z", {without_time: true})}}
Do companies have clear hiring process for Data Scientist role?
We recently started a survey on "Data Science Interviews in 2018". As per the current results, around 40% people think hiring companies don't have clear job role as well the correct hiring process.What's been your experience? BTW, to get the final survey report, please take this survey here.
4
1
Tejas askedon {{::getFormatedLocalTime("2018-04-06T05:38:12.796Z", {without_time: true})}}
Can anyone please help me interpret this box plot?
I understand what box plots are but I am unable to draw any insights from images like this.
5
4
Nipun askedon {{::getFormatedLocalTime("2018-03-27T14:49:43.607Z", {without_time: true})}}
Will demand of data science jobs reduce in longer run?
There is a huge demand for data science jobs in short term (<1-3 years) but I'm afraid that this demand may go down in longer run (>3 years). Reasons:I have observed that after the initial data modelling and implementation, there are no new projects in many companies. For routine improvements, companies may just need a fraction of data scientists they employ right now.Just like web development, ML and data science is seeing modularization and increasingly being packaged as a service (AWS and Google are marching ahead in this). This means there will be fewer data scientists needed to implement same projects in future than today.What do you folks think?
3
1
Vaibhav askedon {{::getFormatedLocalTime("2018-01-10T06:13:55.123Z", {without_time: true})}}
How to detect video language programmaticlly
Hi,I am working on a project where user will upload his/her video from android app. Once it gets uploaded on the server. I  want to detect the langauge of the video programmatically using php. And currently i am looking for free solution i don't want spend money at this stage. How i can achieve this, do i need to perform machine learning or AI on this .  
0
0
Ram askedon {{::getFormatedLocalTime("2017-09-28T07:23:33.571Z", {without_time: true})}}
Big Data and ML algorithms
Would like to know how and what technologies to be used to get distributed data from many servers that to be used in R or Python for ML algorithm.  Assume we have data files stored in distributed servers. How to get those "big" data into R or Python to get Machine Learning (ML) processed?
1
2
Nipun askedon {{::getFormatedLocalTime("2017-09-11T11:06:38.447Z", {without_time: true})}}
Why exactly is Python the most popular langauge for data science use cases?
Stackoverflow says Python is the fastest growing major language: https://stackoverflow.blog/2017/09/06/incredible-growth-python/They don't say why, but a big reason for that growth is apparently due to Python's popularity for Data analytics, data science and BI use cases.What specific features and capabilities make Python so popular again for these use cases?
1
5
Sujeet askedon {{::getFormatedLocalTime("2017-09-08T12:33:32.629Z", {without_time: true})}}
How to develop a Bot from scratch?
I want to build a bot which will show minimum price for hotel booking and flights
1
3
Nikunj askedon {{::getFormatedLocalTime("2017-08-23T07:18:07.124Z", {without_time: true})}}
Reddit's feed algorithm - how can you improve it?
I read about Reddit's feed/recommendation algorithm:  https://medium.com/hacking-and-gonzo/how-reddit-ranking-algorithms-work-ef111e33d0d9This is how it looks like:# Rewritten code from /r2/r2/lib/db/_sorts.pyxfrom datetime import datetime, timedeltafrom math import logepoch = datetime(1970, 1, 1)def epoch_seconds(date): td = date - epoch return td.days * 86400 + td.seconds + (float(td.microseconds) / 1000000)def score(ups, downs): return ups - downsdef hot(ups, downs, date): s = score(ups, downs) order = log(max(abs(s), 1), 10) sign = 1 if s > 0 else -1 if s < 0 else 0 seconds = epoch_seconds(date) - 1134028003 return round(sign * order + seconds / 45000, 7)​Question: Not changing the business rules, are there any improvements you could make to improve the algorithm?
0
0
Priyank askedon {{::getFormatedLocalTime("2017-08-17T10:28:03.612Z", {without_time: true})}}
How to gather trending news on a topic ?
Let's say we have a community that is interested in a topic such as "marketing and growth" or "DevOps". Every day we want to discover trending news for this community. How to do this keeping the relevance high AND with minimal manual curation? Some options we have exploredUse a service like Discovery service from Watson. We tried it but didn't find it much relevant. Is there an alternative like Google news?Follow around 25 Twitter handles or a specific hashtag and analyze thee tweets. Apply some heuristics to check if they contain news and use engagement metrics to guage the quality. Rank all of them to find the winners. (Twitter's engagement may be more on "virality" than "quality")Subscibe to RSS feeds from various blogs (do they still exist?)What'd you try?Any other ideas would be welcome too.
1
1
Neelesh askedon {{::getFormatedLocalTime("2017-05-02T10:53:57.974Z", {without_time: true})}}
Facebook just open sourced its deep learning framework Caffe2, how does this compare to other options?
The deep learning framework can be found here https://caffe2.ai/ since A.I. is very hot right now, how will developers decide which framework they can adopt, given the various options available. 
0
1
RishiMadhavM askedon {{::getFormatedLocalTime("2017-04-24T14:42:56.635Z", {without_time: true})}}
Are we moving towards AI everywhere, will the skills we have currently be obsolete in the near future?
I recently came across an article that had Elon Musk's new startup called Neuralink which suggests that in a couple of years we will have human to human telepathy.
0
1
Nikunj askedon {{::getFormatedLocalTime("2017-04-24T14:21:27.847Z", {without_time: true})}}
What kind of AI will Facebook need to read brain waves?
Read this article and was wondering how will Facebook do this.https://www.bloomberg.com/news/articles/2017-04-19/facebook-s-building-8-envisions-using-brain-waves-to-type-words
0
0
Manas askedon {{::getFormatedLocalTime("2017-04-21T11:37:45.765Z", {without_time: true})}}
Why is R such a popular data science tool?
I've seen a lot of people talking about R in general. What makes it stand out from everything else available in the market right now.
3
1
Neelesh askedon {{::getFormatedLocalTime("2017-04-21T11:43:20.086Z", {without_time: true})}}
How can R and Hadoop be used together?
 
1
2
Rajat askedon {{::getFormatedLocalTime("2017-04-22T14:43:43.224Z", {without_time: true})}}
How will you redesign the recommendation system for LinkedIn?
LinkedIn's feed is becoming messier with time. As a data scientist, how will you build a better recommendation system?
0
0
Nipun askedon {{::getFormatedLocalTime("2017-04-19T13:19:41.636Z", {without_time: true})}}
At the MVP stage in a company, data is not abundant. How does one go about making machine learning models at this stage?
At the MVP stage where data is not available in abundance, how do you cope up with analyzing and predicting data. If you are able to build and predict with this small dataset, will the algorithm still hold relevance or will there need to be changes when there is a sudden inflow of complex data points. 
3
2
Manas askedon {{::getFormatedLocalTime("2017-04-19T10:33:26.660Z", {without_time: true})}}
Why would you choose Tensor Flow over Theano?
  
0
1
We use cookies to ensure that we give you the best experience on our website. If you continue to use this site we will assume that you are happy with it.
Okay
Privacy policy
File upload not supportedAudio recording not supported
This browser does not support file upload. Please follow the instructions to upload your resume.This browser does not support audio recording. Please follow the instructions to record audio.
  1. Click on the 3 dots
  2. Click on "Copy link"
  3. Open Google Chrome (or any other browser) and enter the copied link in the URL bar
Done