The AI recommendation functionality is exclusive to accounts with a Teyuto Prime plan.
Understanding Teyuto's AI Recommendation Algorithm
Teyuto's AI recommendation algorithm leverages user events to power its recommendations. This guide provides an overview of how the recommendation algorithm works and how developers can manage events through Teyuto's APIs to optimize recommendations.
Overview of Teyuto's Recommendation Algorithm
The recommendation algorithm in Teyuto is designed to personalize content for users by analyzing their interactions and behaviors. The algorithm processes various types of events, such as views, likes, comments, and shares, to understand user preferences and suggest relevant content.
Key Components
- Event Tracking: User interactions with content are tracked as events.
- Event Processing: Events are processed to update user profiles and content metadata.
- Recommendation Generation: Based on processed events, the algorithm generates personalized content recommendations.
Managing Events via Teyuto's API
To effectively utilize Teyuto's recommendation algorithm, developers need to manage events through Teyuto's API. Here’s how you can do it:
1. Setting Up the API
Before you can start sending events, ensure that your application is correctly set up to interact with Teyuto's API. Obtain your API key and ensure you have the necessary permissions.
2. Sending Events
The following table lists default events natively handled by Teyuto, a system integrating Artificial Intelligence (AI) to analyze and assess user interactions. These events, represented in the type
column and associated with an AI rating
score ranging from -1 to 1, reflect various user actions on the platform.
It is important to note that the default events listed here cannot be overridden or modified by the user. However, to enable the management of custom events without conflicts with native ones, Teyuto uses parameters named
custom_{type}
through APIs. These parameters allow developers to create and manage custom events, providing flexibility and adaptability to specific needs without interfering with the logic of default events.
Type | AI Rating |
---|---|
user_created | null |
user_session | null |
user_login | null |
user_edited | null |
content_searched | null |
collection_searched | 0.3 |
video_searched | 0.3 |
user_delete | null |
category_viewed | 0.1 |
tag_viewed | 0.2 |
collection_favorite | 0.5 |
collection_viewed | 0.3 |
video_played | 0.5 |
video_stopped | null |
video_completed | 0.7 |
video_favorite | 0.7 |
abandoned_cart | null |
collection_share | 0.2 |
subscription_created | null |
payment_created | null |
subscription_canceled | null |
Example: Sending a View Event
To send a view event, make a POST request to the Teyuto API with the relevant event data.
API Documentation: Create Event
curl --location --request POST 'https://api.teyuto.tv/v2/events' \
--header 'Authorization: YOUR_API_KEY' \
--header 'Content-Type: application/x-www-form-urlencoded' \
--data-urlencode 'type=custom_event_1' \
--data-urlencode 'user_id=1234' \
--data-urlencode 'rating=0.5' \
--data-urlencode 'date_created=YYYY-MM-DDTHH:mm:ss'
3. Processing Events
Once events are sent to Teyuto, they are processed to update user profiles and content metadata. This processing helps the recommendation algorithm to understand user preferences and behaviors better.
4. Retrieving Recommendations
To retrieve personalized recommendations for a user, make a GET request to the Teyuto API.
API Documentation: List of Contents
curl --location --request GET 'https://api.teyuto.tv/v2/contents' \
--header 'Authorization: {USER_TOKEN}'
The response will contain a list of recommended content based on the user's event history.
Note: The algorithm may take some time to become effective after the initial data is provided. It requires a sufficient amount of event data to accurately model user preferences and behaviors.
Best Practices
- Consistent Event Tracking: Ensure that all relevant user interactions are consistently tracked and sent to the API.
- Timely Updates: Send events in real-time or at regular intervals to keep user profiles up-to-date.
- Data Accuracy: Ensure the accuracy of event data, including correct user and content IDs, and accurate timestamps.
Conclusion
By effectively managing events through Teyuto's API, developers can enhance the performance of the AI recommendation algorithm. Consistent and accurate event tracking enables the algorithm to provide more relevant and personalized content recommendations to users, improving their overall experience on the platform.
For more detailed information on Teyuto's API, refer to the official API documentation provided by Teyuto.