Universal Models

Default model ensemble aiming to work well in most common use cases. It is an ensemble of collaborative filtering and content-based recommendation, backed by popularity-based models when there's not enough data in the given context. The ensemble is being continuously optimized by AI so it automatically adapts to incoming data.

Universal model ensemble suitable for putting recommendations on the homepage / welcome-screen scenarios.

Universal model ensemble suitable for user-based recommendations reflecting the personal tastes of the current user.

Universal model ensemble suitable for either recommending items which are similar to the given item (in case of Items-to-Item recommendation), or recommending users who are similar to the given user (in case of Users-to-User recommendation).

Universal model suitable for recommending globally popular items, taking into account all interaction types. More important actions (purchases, cart-additions) have higher weight for computing the popularity than less important actions (detail-views).

Universal model ensemble suitable for recommending globally popular segments.

Model returning the list of recently viewed items for a particular user.

Model recommending similar items based on visual similarity (using deep-learning neural networks to process the images provided through the image properties). While the image similarity can take part in other Logics as well, in recombee:visually-similar, it is explicitly used as the primary model. Image processing may take some time. Therefore, changes in image item properties do not take effect in real-time. The functionality is still in beta mode.

Model recommending similar items based on item properties (numerical values, sets, and texts). While property values can take part in other Logics, in recombee:similar-properties, they are explicitly used as the primary data source for recommending similar items or uses. Various machine-learning methods, including NLP deep-learning networks that process unstructured text descriptions, are used to compute the similarities.

Model ensemble suitable for recommending items in periodic e-mailing campaigns. It contains fine-tuned rotation settings so that the set of recommended items for a particular user changes with each e-mail sent.

By default, rotation only takes into account recommendations (e-mails) at most 14 days old. Should a longer time period be needed for your use case, kindly contact support@recombee.com.

Model suitable for recommending items ordered by a ReQL expression given as a booster

Recommends items that are contextually related to the Item Segment and should be of interest to the user. Items may belong to the Segment or be outside it if they are strongly relevant.

Consider using recombee:personal-from-segment if you want recommendations strictly limited to items within the Segment.

Examples:

• News: For the "Climate Change" Segment, recommend articles about renewable energy (inside the Segment) or related topics like electric vehicles (outside the Segment).
• Music: For the "Jazz" Segment, recommend jazz-related songs, including classic tracks (inside the Segment) or modern experimental pieces (outside the Segment).
• E-commerce: For the "Digital Cameras" Segment, recommend products like digital cameras (inside the Segment) or accessories like tripods, lenses, or lighting kits (outside the Segment).
• Video Streaming: For the "Science Fiction" Segment, recommend sci-fi movies (inside the Segment) or related dystopian thrillers (outside the Segment).

Recommends items from the Item Segment, tailored to the user's preferences.

Consider using recombee:related-to-segment if you don't need all the recommended items to belong to the Segment.

Examples:

• News: For the "Environmental News" Segment, recommend articles on climate policy or renewable energy that align with the user’s reading history.
• Music: For the "Jazz" Segment, recommend jazz songs matching the user’s listening habits.
• E-commerce: For the "Digital Cameras" Segment, recommend digital cameras the user is most likely to be interested in, based on their browsing and purchase history.
• Video Streaming: For the "Science Fiction" Segment, recommend sci-fi movies that match the user’s preferences.

Recommends the most popular items within the Item Segment, based on overall engagement. Use this Logic to highlight trending or widely appealing items.

Examples:

• News: Most-read articles from the "World Politics" Segment.
• Music: Most-streamed tracks from the "Rock" Segment.
• E-commerce: Best-selling smartphones in the "Smartphones" Segment.
• Video Streaming: Most-watched action movies in the "Action Movies" Segment.

This Composite Logic provides a comprehensive recommendation in a single response consisting of:

• Top Segment for the User: It identifies and returns the most relevant Item Segment (e.g., the top category) for the user from the specified Segmentation.
• Top Items from the Segment: Along with the Segment, it also returns a list of the top items within that Segment.

This Logic is ideal for recommending a section and its content together, making it particularly useful for personalized homepage sections.

If you want to display multiple distinct sections in a personalized order on your homepage, you can achieve this by making multiple Composite Recommendation requests within a single Batch call.

For more specific use cases, such as "Videos From The Top Genre For You" or "Products from the Top Category For You," there are specialized Logics available.

This Composite Logic delivers two results within a single response consisting of:

• Item the User Interacted With: The Logic selects and returns an item that the user has interacted with, based on the interaction types specified by the interactionTypes parameter (e.g., Watched, Purchased, Liked).
• Top Related Items: Simultaneously, it returns a list of top items related to that specific interacted item.

This Logic is ideal for general "Because You..." scenarios, where recommendations are tailored based on the user's previous interactions. For more specific use cases, such as "Because You Bought" or "Because You Watched", there are specialized Logics available.

This Logic is designed for infinite swiping feed experiences (such as reels or short-form video feeds), where users continuously browse content and can express feedback through likes or dislikes.

It optimizes recommendations in real time based on a configurable combination of user engagement (likes, number of viewed items, purchases), and time spent consuming content.