Posteye

The quicker, the better.

Comment

Inspiration

After purchasing an item from an online store and didn't like it. A post on Facebook asked the people who purchased the item for their opinion. There were alot of comments both good and bad about the item. That kept me asking, how do the page managers cope up with that.

What it does

Posteye organizes the different comments made by the page users into different categories depending on the user's Sentiment's, users emotions and behavior so that the post manager easily gets a clear picture of the customer's reaction. The post manager can decide how to use the data provided to him by Posteye. For example, The Post manager can like, reply or just interpret the graphs generated from the expert.ai data.

How we built it

I built it using flask framework to provide an html/JavaScript/css frontend which captures users input and sends it using ajax to the python backend which gets the sentiment's, emotional and behavioral traits using the expert.ai python api after getting all the comments on the post using the Facebook graph python api. The data obtained is sent back to the html/JavaScript/css frontend which draws the graphs using chart.js api to give the user a visual representation of the data obtained.

Challenges we ran into

I was having a hard time interpreting expert.ai data... though it turned fun.

Accomplishments that we're proud of

I was able to merge the data obtained from expert.ai with the data obtained from facebook graph.api into a smart filter.

What we learned

I learned alot about the interpreting the expert.ai returned data.

What's next for Posteye

*Private Replies

*Graph prediction and description.

*Addition Of Other Social media platforms.

Built With

Share this project:

Updates