This is an endpoint for api for all our hotel ranking scripts.
Information | Value |
---|---|
Docker Image | kaenova/traveloka-hotel-ranking |
Port Open | 8001 |
Variables | Description |
---|---|
DB_USER | Database Username |
DB_PASSWORD | Database Password |
DB_HOST | Database IP Address or Host name |
DB_PORT | Database Port |
DB_NAME | Database Name |
DB_CON_STRING | Database Connection String (if used, will ignore all DB ENV above) |
Hotel similarity endpoint based on almost all columns except "name", "neighborhood", "type_nearby_destination", "image_links", "id"
.
Endpoint | Input | Output |
---|---|---|
POST /hotel-similarity/{hotel_id} |
- | List of Hotel ID (Currently hardcoded to give 10 recommendation) |
If Hotel Ids is not cached in ML Endpoint or Cannot be Processed for Hotel Similarity:
A recommendation system based on weighted similarity sentiment reivews
Endpoint | Input | Output |
---|---|---|
POST /sentiment-similarity/{user_id} |
- | List of Hotel ID with number hotel recommendation ranged from 0 to 10 |
Please Be Aware that the recommendation is not always give 10 ids of hotels.
Example:
If Success:
If Success but No Recommendation Because of all hotels is been reviewed by the user:
Trending Recommendation endpoint is based on the number of reviews and its average rating with the formulation of then end score: number of reviews * (average rating * 1.2), the output would be a list of Hotel ID sorted from having the best to the worst score
Endpoint | Input | Output |
---|---|---|
POST /trending_system/ |
- | - |
Near You API endpoint is based on user's position, and it will show several hotels that are near the user's location.
Endpoint | Input | Output |
---|---|---|
POST /trending_system/ |
- | - |
If you want to force caching in the ML endpoint, just hit this method.
Endpoint | Input | Output |
---|---|---|
POST /re-cached/ |
- | Text |
If you want to check how many rows cached
Endpoint | Input | Output |
---|---|---|
GET /rows-cached/ |
- | Dictionary |
All system error on ML sides will give you a 500: Internal Server Error
status.
CC22-HO01 ML Teams.