Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[Poster Session] Contents #119

Open
jang1suh opened this issue Dec 14, 2019 · 6 comments
Open

[Poster Session] Contents #119

jang1suh opened this issue Dec 14, 2019 · 6 comments

Comments

@jang1suh
Copy link
Contributor

No description provided.

@jang1suh
Copy link
Contributor Author

jang1suh commented Dec 14, 2019

What is Triplannet?

Triplannet is a service that helps users to organize, search, and share plans for a trip. Our service provides a simple drag-and-drop UI when users build plans, and this can reduce users’ inconvenience greatly. By adding another user as a collaborator, you can edit a travel plan together with fellow travelers. Triplannet also provides a community for travel lovers where they can share their successful trip plans. Take a tour of Triplannet community when you are planning to travel somewhere. Your travel plan will be much more fruitful!

@jang1suh
Copy link
Contributor Author

jang1suh commented Dec 14, 2019

Main Features

  1. Create travel plans: with drag-and-drop UI
  2. Add another user as a collaborator of a travel plan: edit individually & merge
  3. Share travel plans with other users around the world
  4. Recommend travel plans for user: Contend-based recommendation (ML applied)
  5. Fork other user's plan and edit: don't have to start from scratch.
  6. Search travels with #tags

@jang1suh
Copy link
Contributor Author

jang1suh commented Dec 14, 2019

Recommendation Method

Recommendation feature is targeted to user who want to look around the plans. We recommend next plan to look. Our idea was hybrid method of content-based recommendation + collaborative filtering with implicit feedback.

  1. Text content based - Gather the important text from plan and calculate embedding vector. Recommend by similarity of vector
  2. Block distribution based - Block distribution represent the characteristic of plan. Plan with similar block distribution will be recommended.
  3. User activity based - Recommend by user activity like views, likes, comment by CF model with implicit feedback.

Lack of User activity data, 3rd method didn't work well, so we apply 1,2th method

@jang1suh
Copy link
Contributor Author

jang1suh commented Dec 14, 2019

System Design & Architecture

Backend model #63
Each modification of a travel plan is stored as a form of "Commit"

@jang1suh
Copy link
Contributor Author

Service Demo

screenshots

@jang1suh
Copy link
Contributor Author

Future Work

content

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant