BizFeed
An online platform what provide feedbacks of limited distribution of videos through Face Recognition and Sentiment Analysis.
Project developed for GSUHACK-SPRING2017
Motivation
Schools, organizations and large business are composed of various departments and students/employees in each department require different trainings. Transferring to another area may require redundant training and sometimes training video quality is very poor and employees are reluctant to provide feedbacks. BizFeed allows orgs to manage all their training videos in one place, and providing feedbacks about how users responses while watching the video, reflecting the quality of video through concentration and sentiment analysis with face recognition technology.
BizFeed can also be an useful tool for independent film/video makers to gather more accurate feedbacks with less bias through technology, and controlling limited distribution of the video.
What it does
Accounts are organization based, and anyone else/regular users are invitation only through the organization. Video clips can be uploaded, managed, and assigned to specific users/user-groups by admin. Real-time data will be collected when user is watching the video, and stored in db for further analysis. Admin can make sure each user finished watching all required videos and can view analysis on each clips, which may provide useful feedbacks for future improvements.
How we built it
Php
JavaScript
d3.js
Affectivia
chart.js
postgres-sql
Apache Http Server
Bootstrap
Html
Css
Challenges we ran into
Sooooooo much...
Learning a new language, new APIs, new packages and making everything work within a limited time frame..
Issue with environment setup for php, and deployment..
DevPost markdown does not support list?
Accomplishments that we're proud of
Project accomplished~
What we learned
LOTS!!
Learned new language(Php), Learned fun APIs, Debugging, Data Visualization
Issue with Chrome browser that it does not automatically update .css files when refreshed..
What's next for BizFeed
Move to Node? Deploy on Aws? Pending further development:)
Log in or sign up for Devpost to join the conversation.