Hello World! We're a team consist of 10 peoples from around Indonesia in a programs called Bangkit, founded and arranged by Google, Traveloka, Goto and Dicoding. We're in collaboration with Traveloka creating a prototype with a theme of optimizing hotel ranking based on business cases creates a prototype of hotel recommendation application.
More Information
A hotel recommendation application built with 5 different types of algorithm to provide you the best experiences to stay in Bali! This projects covers all 100 Bali's Hotel from different residence so that you can enjoy every bit of "Bali Experience and Culture." Built with low-cost cloud infrastucture so that the maintenence of this applicaiton is relatively easy to scale up, and also built with Android Kotlin for the optimize and native performance so your experience searching for the best hotel becomes more easy!
Our teams formed 3 different division with each division has a different tasks. From Cloud Computing which provides an API for the data and connecting Machine Learning Division with the Mobile Development Division. Machine Learning Division which will provide an algorithm to optimize what and how the hotel ranking displayed. Mobile Development Division which makes an application interface for display the hotel ranking prototypes.
Bangkit ID | Name | Social Media |
---|---|---|
C2306G2618 | Christopher Alvin Buana | LinkedIn Instagram |
C2144G1639 | Freddy Andreas Chandra | LinkedIn Instagram |
C2306G2621 | Michael Georgisva | LinkedIn Instagram |
Bangkit ID | Name | Social Media |
---|---|---|
M2012F1254 | Bagja 9102 Kurniawan | LinkedIn Instagram |
M2306G2624 | Farel Arden | LinkedIn Instagram |
M2011F1179 | Jessica Jesslyn Cerelia | LinkedIn Instagram |
M2012f1247 | Kaenova Mahendra Auditama | LinkedIn Instagram |
Bangkit ID | Name | Social Media |
---|---|---|
A7279f2409 | Jaka Asa Baldan Ahmad | LinkedIn Instagram |
A2005f0430 | Kiki Abdullah | LinkedIn Instagram |
A2012f1257 | Zendy Bramantia Alfareza | LinkedIn Instagram |
- Sentiment-Prediction-Modelling. Notebooks of our research and modelling for Sentiment Prediction on Hotel Reviews.
- NER-Location-Modelling. Notebooks of our research and modelling for Named Entity Recognition (Location Only).
- For-You-Pipeline-Modelling. Notebooks of our research and modelling of "For You" Page.
- Hotel-Ranking-Calculation-Endpoint. A stateless endpoint of our system to calculate and filter hotel rankings.
- Sentiment-Prediction-Endpoint. An endpoint for our system of predicting hotel reviews sentiment.
- Location-Recognition-Endpoint. An endpoint for our system to improve search engine.
- DL-Hotel-Rating-Estimation-Endpoint. A stateless endpoint of our system to estimate users hotel rating using deep learning approach.
- Hotel-Recommendation-Research. A research of our recommendation systems.
- hotel-ranking-app. Hotel Ranking App based on Android Platform with Kotlin.
Keep Bangkit Spirits Up!
C22-HO01 Teams.