This is the React JS Frontend of the Social Media Image analysis project. This makes requests to the Vision-API.
The purpose of this project was to create a tool that could quickly analyze images of a public social media account. Since looking through hundreds of images is time consuming and subjective to opinions, each of the images from a social media account were processed with Google Cloud's Vision API. From the responses, labels are accumulated, filtered, and tracked to build textual data that can represent a person's interests, activities, or hobbies. Individuals' online personas can also be compared to determine if people share common interests. This project idea can be used in many different applications: suggestion algorithms for friends or pages, dating app matching, job recruiting, and even social media marketing analysis.
yarn
yarn start
This is the Home Screen where you can pick which social media to retrieve images for. You can also navigate to the Compare and Analytics Components.
Here you can see the Vision API labels of a specific user and a specific platform. The bar chart is interactive and is implemented with a React variant of Chart JS.
Here you can compare two different usernames. The username url parameters query MongoDB and show labels in common and their respective counts.
This is where you can see how many unique users, labels, and images that are stored in MongoDB.
Components are a little messy right now. Will clean up eventually.