Docs
User Guides
Recommender
How it Works

ecosystem.Ai Recommenders

The ecosystem.Ai recommender capability uses sophisticated tools that enable you to build more effective offers (recommendations).

Worker architecture allows for use of the latest modeling packages. Architected from the ground up for durability and performance, it returns results in real-time and is highly resilient. You also have easy deployment to production at the push of a button.

There are many ways to structure your recommenders:

  • Multinomial models
  • Collections of binomial models
  • Incorporating an exploration component
  • Stacked ensembles of recommendations.

The Value of Real-Time Recommendations

Real-time allows you to use the latest information about your customer. Letting you use real-time features such as location, time of day, balance, etc. which aren’t available to batch models at all. Real-time ultimately removes the problems associated with batch-generated offers becoming irrelevant over time.

The difference between batch and real-time extends beyond the recommendation itself, in real-time, you will be able to view, analyze and adjust according to the customer response, the very moment it happens.

Recommender Structures

There are a number of different model frameworks you can use in your recommender structure:

  • Multinomial models:

A single model is built which predicts the offer that a customer is most likely to engage with.

  • Collections of binomial models:

A model is built for each offer, which predicts the likelihood of a customer engaging with that offer.

  • Incorporating an exploration component (Bandit Algorithms):

Rather than only presenting an offer which is predicted to be the most likely to be engaged with, occasionally present other offers in a way that allows you to explore whether the human behavior in your system has changed.

  • Stacked ensembles of recommendations:

The output of one model can be used to inform the next model. For example, one model could recommend a design construct, while the next model sets the messaging within that design.

  • Knowledge-Based Systems:

Recommends offers based on specific knowledge about users and items, often using a rule-based framework.

  • Reinforcement Learning:

Models the recommendations as a series of sequential decisions, the uses reinforcement learning algortihms to optimize long-term interactions.

🪶
Some more common model frameworks
  • Collaborative Filtering
  • Content-based Filtering
  • Hybrid Methods
  • Deep Learning Models