Insanely Analytics

The space for feeding one's crave on analytics

My new R Shiny Web Application – “Prediction of Success of a Movie with Twitter Corpus” Check it out and spread the word!

Check out my new Shiny web application designed with R (majority), Shiny, tailored CSS and HTML5. With this application, you can analyze sentiments of not just movies but also a brand, a product or a personality. It uses tweeple’s post(tweet) as its data set. It’s been trained with “Naive Bayes Model” attaining 86% accuracy having trained for 10K rows of texts. You can also visualize the location of tweets with this application that uses interactive, dynamic Open Street Maps(OSM). Its been designed to suit well on smartphones too.

Click here to get redirected to the application

ss 5

I would love to hear your valuable feedback.

 

Performance  analysis of my Naive Bayes Model vs. Existing models on Naive Bayes

Untitled.png

I’d like to express my profound gratitude towards my data science tutors, one Mr. Vladimir Kirilenko and Mr. Ali Santacruz. Without them this work of mine would’ve been just a night mare! Thank you once again sir @Vladimir and @Santacruz! Your timely help and guidance cannot be calibrated.

Advertisements

9 thoughts on “My new R Shiny Web Application – “Prediction of Success of a Movie with Twitter Corpus” Check it out and spread the word!

    1. My apologies @Gautam ! Users have consumed almost 20 hours out of 25 active hours provided. So, I’ve put it down as I need to run it for my project demonstration purposes. Will revert to you once it is open.

      Like

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: