Last spring semester, I developed a SNS application called PICKME! for indecisiveness with my teammates as a project in the class of KAIST CS408 Computer Science Project. This project aimed to help people who want to resolve their indecisive issues by asking to massive users who experienced similar issues or just asking their opinions via votes or comments.
“How can we resolve their issues effectively?”
Our goal was the developing a mobile service application for end users and recommender system to recommend personalized contents so that users can easily find their similar or interested issues and resolve their indecisive issues which they might be interested.
Key Features
1) Post a card that user want to share to resolve his/her issue
2) Participate vote to share his/her opinions
3) Communicate via comments to share his/her thought
4) Explorer and get recommended contents from recommender system
We used Unity for a client application for Android and iOS platforms, and Firebase and Django for database and server networking. We also extracted and analyzed 15,000+ tweets to test our recommender system based on collaborative filtering with LDA topics. To measure our project, we did a survey to KAIST members with the mobile application as a beta testing.
References
[1] Paul Covington, Jay Adams, Emre Sargin, “Deep Neural Networks for YouTube Recommendations”, in 2016
[2] CARLOS A. GOMEZ-URIBE and NEIL HUNT, Netflix, Inc., “The Netflix Recommender System: Algorithms, Business Value, and Innovation”
[3] Michael Hahsler, “recommenderlab: A Framework for Developing and Testing Recommendation Algorithms”
[4] P.Selvaperumal, Dr.A.Suruliandi, “SHORT MESSAGE CLASSIFICATION ALGORITHM FOR TWEET CLASSIFICATION”, in 2014 International Conference on Recent Trends in Information Technology
Seungwook Chi 