50 AI engineers have collaborated to analyze children’s learning applications’ data and have built a recommendation system that adapts to the level of the children as they progress in the app.

The problem

With a personalized study plan, Poikilingo works on developing language skills and cultural awareness for bilingual children, including immigrants and refugees, who need to learn the language of the host country as fast as possible to be able to start in the school system.

The goal of this challenge was to build a machine learning model that adapts to the data and the children’s progress in the applications, ensuring a personalized learning experience for each individual child.

The project outcomes 

The team built a personalized recommendation system where the child’s study plan adapted to the level of the child as she progressed in the plan by collecting, analyzing, and inspecting the data.

The model synchronized with the child’s progress within a given interval and improved the machine learning model. The team trained the model to identify the child’s next steps based on the child’s progress in the study plan. Additionally, they created a custom dashboard to analyze, collect, inspect, etc., all the data users (inputs), all the machine learning data, and the outputs.