- Integration of peach collection libraries in the website or in iOS and Android mobile apps
- Tracking play start if the content is a video or audio
- Acquire metadata for the content (title, category, tags, keywords)
- Track if content was recommended
Implement the algorithm
For starters, this is typically done by the PEACH team!
- Explore data and define which algorithm is suitable
- Write the algorithm in PEACH Lab
- Create a Task to automate the creation and update of the model
- Create a REST Recommendation APIs endpoint to deliver recommendation to clients
Evaluate the algorithm
- Use PEACH Lab for offline metrics based evaluation
- Use Spectrum to see recommendation visually and to present results within your organization
Display recommendations for users
The recommendations need to be made available to the end users. This can be in the form of a list on a web page, the automatic launch of a video after watching or a sidebar on a particular page, etc. The strategy on what to display and where to display it varies based on the product delivered to users, the editorial line, the UX etc.
- Create the UI components to display the recommendation
- Integrate retrieval of recommendation on website
- Integrate fallback of trending when no recommendation is provided
Below are two examples of how the integration of the recommendation has been implemented on the RTS and RTP websites.