Emofilia

Inspiration

The emotion-based song suggestion algorithm & web scrapers served as our source of inspiration. We believed that, we could create a mixture of both and convert them to something very fruitful using Cohere.

What it does

As we see many quotes poster uploaded on social media over the internet. So this project will categories this quotes based on the emotions of the content. It will also extract the content from the post which can be linked with the post or image so that it will be also helpful to reach out that particular post based on the content.

How we built it

We have used a Tweet's dataset which content different categories of tweets, and we have used CohereAI online NLP platform to build our classification model for different categories of posts. We have used EasyOCR along with Computer Vison to extract the texts from the images. So the program takes an image as the input and extracts the text from the input and run the classification model on the extracted text to categories the text and then return the Content of the post along with its Category

Challenges we ran into

We had to learn about Cohere from scratch also we faced many issues during handling the platform but the tech support was really helpful and got us out of every problem we faced. We faced some issues with WIX by velo even though our site was ready we found no options to deploy our model and tech support wasn't that responsive.