Crossing Minds is democratizing access to recent gains in artificial intelligence research by building the world's best content recommendation platform.
Leverage the Best Recommendation API.
Crossing Minds is creating a universal recommendation API that can be integrated into your business with minimal effort. We implement incredibly efficient algorithms that eliminate the recommendation guesswork for current users and when onboarding new users.
Our experienced team walks you through the process of data collection and implementation so you can get a solution unique to your business and customers. We focus on recommendation accuracy while protecting the privacy of user data.
What do you need on your side to get the best recommendation?
The only data REQUIRED to train the best recommendation engines for your business is, of course, the interactions the users have with your products. You can, of course, upload those interactions in 2 different way:
Ratings or Explicit Feedbacks: Explicit feedbacks clearly describes on a fixed scale the rating a user gave to an item (like/dislike, star rating, etc.). For more information about how to prepare your file and preprocess them, please read our API documentation!
Direct User Interactions: or what we call "Implicit Feedback." Those are all the interactions you can find, for instance, in your Google Analytics, the clicks, the scroll, the page view, etc. User Interactions represent different interactions a user may have with an item, often hints whether the user likes or not an item.
Naturally, he more information you can provide about your consumer (while respecting privacy and security) can help the recommendation engines build your consumers' DNA.
Sending user data is OPTIONAL but can improve the recommendation quality, especially when it comes to cold-start recommendations where your user hasn't interacted with your products yet.
Enriching the information our algorithm can have about your items is strongly recommended, although also optional!
Using rich properties for your items gives two advantages:
1 - It improves the recommendations, especially for both cold-start problems where the algorithm relies only on properties (such as Semantic Graph Embedding from genres and tag, or Deep Content Extraction from text and images)
2 - It enables you to dynamically filter the recommendations on items satisfying certain criteria (such as a price smaller than a threshold given at runtime)
What do you get out of this??
Given an item ID, the API will return similar items than the one visited.
Under the hood, the similarity is computed using a hybrid model that leverages both the content-based part (the item properties) and the collaborative-filtering part (the interactions).
SESSIONS BASED Recommendation
This feature is ideal when you need to generate recommendations for new users.
We are talking about first time visitor or people who have not already signed up in your application. User that you have to convince right now!
Given a list of interactions for an anonymous session, the API will return the best items.
Given a user ID, the API will return the very best items that match the profile of each specific user.
This is the most common endpoint to get recommendations for users that are already in your database.
See Under the Hood to read on the underlying technologies.
All our recommendation endpoints support filters parameters specific to your business.
Using recommendation filters, you can return specific items satisfying given criteria on their properties, such as price smaller than a threshold, or having a certain tag.
Interested in trying our Recommendation API for free?
We are very exited to share our newly released Recommendation API, which businesses can utilize to offer their customers the best recommendations possible.
If you are interested in being among the first to try it out, please answer the interest form. Our team will reach back to you as we are processing the flow of applications. Thanks!
How to make the best of our Recommendation API
Cold Start Problem
For a first time user of your brand, attempting to provide recommendations without data-sets reflecting interests and tastes can be challenging, this is what we would categorize as a cold-start problem.
Our API has been build with this problem in mind, and can help you being relevant to any new users, concretely increasing your short-term retention
Cross Domain Recommendations
We are able to build a “digital DNA” through the embedding of information gathered cross-platform by creating an anonymous mathematical representation of a consumer that contains all their tastes and behaviors. This gives any company an obvious competitive advantage and can further be leveraged for numerous predictions and analyses such as recommendation engines, churn prediction and predictive lead scoring.
Because of our improved embeddings, marketing teams can perform segmentation and audience targeting which provides a more accurate idea of the outcome or necessary strategy adjustments regarding specific campaigns, resulting in better customer understanding, better marketing insights and reduce media costs.
We work to make it real!