Skip to content

BinniesLite/Sentiment-Insights-

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

5 Commits
 
 
 
 
 
 

Repository files navigation

Sentiment-Insights

If there is no customer there is no product

Insights from customer feedback.

Shine brighter than your competiton.

The San Juan Mountains are beautiful!

There not enough words in this world to describe how important customer is so for startup and middle size companies to prosper it is essential to understand your customer reaction to the product, customer service, speed of operation


Inspiration ☀️

There not enough words in this world to describe how important customer is. This true especially for startup and middle-size companies to prosper it is essential to understand your customer reaction to the product, customer service, speed of operation

What it does 🤔

Our business tools offer precise insights from customer feedback that allows you to elevate your business, eventually to the moon! Our state of the art ML Models, which deliver 98% accuracy to its predictions, can save you months worth of time and hundreds of dollars. It can produce charts to help visualization and point the businesses in the right direction. Isn't that awesome?!?

  • Understand your prospect and potential customers 😄
  • Easy to use 😵‍💫
  • Save Time ⏲️
  • Save thousand of dollars 💰

How we built it

With PASSION and LOVE The San Juan Mountains are beautiful!

Just kidding, we use Python and it's friends like streamlit, pandas, nltk, matplotlib, numpy, catboost, Scikit Learn. We also add some Reactjs, material ui and Boostrap to make the project looks more colorful 🌈

Challenges we ran into

In Python 🐍

  • We have the most trouble debugging with some of the most noticable errors are when running the counter loop the code doesn't run as expected.
  • The data source is another source of problems since it takes a while for the program to learn between positive and negative comment. It used to product the result with less than 70% accuracy In react 🖥️
  • The UI is not responsive as expected this problem is overlook until there is not enough time to fix it
  • Wasn't able to exchange data between Python and React so the web page main use was changed from a demonstration web page to a display web page
  • It takes too long to create the overlay and finding an image for the background
  • The form is also incomplete thus not included in the final products

Accomplishments that we're proud of

We're proud to say that the AI aka our intelligent AI model with 98% accuracy in determining customer feedbacks Improve data source from last Hackathon. We also go above and beyond the original plan with creating a pie chart that dissect and analyze the customer feed back and produce a pie chart to user. The web page is also very appealing and modern.

What we learned

What's next for Sentiment Insights

We are hoping to demonstrate our product to the user live by communicating data between the front end and the AI will deconstruct the feedbacks and determine if the feedback is a positive comment or a negative comment. With more time and maybe funding we can set up an API page that business only need to send a bulk of data to the api and then we will return it with chart that help them summerize all the things they need

About

Insights from customer feedback.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages

  • Python 100.0%