Sentiment Analysis by LSTM model

• 

  Sequence of words inputted by user

• 

  Resulting prediction and score

Inspiration:

I have a keen interest in human languages and deep learning algorithms, so I wanted to build this program to explore ways to test machine learning capabilities.

What it does:

Users can input a sequence of words and this python program can analyze and predict the sentiment of the sentence (positive, negative, or neutral). The program returns a number between 0 and 1, where numbers closer to 0 demonstrate negative sentiment and numbers closer to 1 demonstrate positive sentiment.

How we built it:

The program was built in python using keras and LSTM model to train a sentiment classifier. It was then deployed to a flask local host server and the webpage was styled using CSS/HTML.

Challenges we ran into:

When attempting to deploy the program onto the host server, there were many errors, and it was difficult to determine what was causing the errors.

Accomplishments that we're proud of:

We made it.

What we learned:

We learned how to use flask.

What's next for Sentiment Analysis by LSTM model:

We want to try to use a Google app engine or AWS to deploy the program onto a public website.

Built With

• css
• flask
• html
• keras
• python
• tensorflow