Recombee is a Recommender as a Service that leverages real-time data to provide the most accurate content and product recommendations for each user.
The recommendations are computed using various machine learning algorithms (collaborative filtering methods, text mining, deep learning, etc.).
The recommendations are updated in real-time - meaning that right from the first click and any following interaction (such as view or bookmark of a product) the system retrains itself and takes this information into account while generating recommendations.
Recombee solution is highly scalable and can serve thousands of recommendations per second.
It can be used for various use-cases in multiple domains. For example, content recommendations can be applied in VoD, news, articles, media, real estate, job portals or gaming. Product recommendations are typically used in e-commerce or classifieds
How Recombee AI works
Recombee provides an intuitive API for sending data and requesting recommendations with SDKs for various programming languages.
For even smoother integration Recombee also provides a WYSIWYG editor of the recommendation widgets in the Admin UI.
After the initial integration, the recommendations can be configured in the Admin UI to determine where on your website where the recommendations will be visible to your users. You can also set the behavior of the recommendation model and configure various filters and boosters with no further changes to your source code.
Get more out of Recombee AI with Segment
Using the Recombee Destination in Segment can further simplify the integration of Recombee into your website or app as the Destination can send the interactions of your users (views, purchases, plays, etc.) to Recombee.
Therefore the only steps left to make the recommendations running at your site are synchronizing the items catalog (using an SDK or by setting a product feed in the Admin UI) and showing the recommendations to your users (using the WYSIWYG editor in Admin UI or by using an SDK).