Sentiment Analysis of Customer Reviews

Inspiration

Inspired by the configuration of the brain, sentiment analysis algorithms imitate how the human brain processes data through an artificial network of neurons.

What it does

Sentiment analysis is a machine learning tool that analyzes texts for polarity, from positive to negative. By training machine learning tools with examples of emotions in text, machines automatically learn how to detect sentiment without human input.

How we built it

We used 8000 customer reviews of different smart phones and later we cleaned the data and performed EDA and used libraries to detect the reviews as positive or negative depending on score of sentiment like if the score is above 0.6 it is considered to be positive and lesser than that will be negative. We used Logistic regression ML algorithm and achieved an accuracy score of 93%.

Challenges we ran into

The challenges we faced are we scraped our own data from amazon and used library Beautiful soup also it's not a easy task to scrape 8000 reviews from online website.

Accomplishments that we're proud of

We achieved an accuracy score of 93%.

What we learned

We learned how can we use real world data to achieve so that brings great insights to service providers to improve their products or whatever they're providing to the market.