RSS
Логотип
Баннер в шапке 1
Баннер в шапке 2
Project

ProgKids Online Programming School Implements AI-Based Learning Quality Analysis System

Customers: ProgKids (Progkids)

Moscow; Education and science

Product: Artificial intelligence (AI, Artificial intelligence, AI)

Project date: 2023/08  - 2024/02

2024: Creation of the MeetProgKids platform

The online school programming ProgKids , as part of a state grant, presented the development of a system to artificial intelligence monitor the quality of the classes. The company announced this on March 4, 2024.

On the specially created MeetProgKids platform, artificial intelligence uses a video and audio stream to analyze the emotional state of a child during a lesson. The pre-trained SpeechBrain model allows you to study in detail whether the student was cheerful and cheerful, or, on the contrary, gloomy and thoughtful. During the lesson, you recognize the face and determine the direction of the gaze. You can solve this problem using special libraries, for example, Pygaze, or by training your own model.

File:Aquote1.png
The collected statistics show that the worse the student's mood during classes, the higher the likelihood that he will leave the course ahead of schedule, "said one of the system developers, Akulov Dinis.
File:Aquote2.png

The quality of the material passed is also recorded in audio decoding, which helps to control the teacher's work and the student's progress. Artificial intelligence tracks the passage of all stages of the lesson: announcement, repetition of past lessons, new material, answers to questions and reflections. This project allows you to focus on an individual approach to each student and guarantees the safety of the child in classes.

How does it work? For each word from the child's speech, a check is made for attitude to one of the three lists (rude, polite and parasite words). Lists were compiled based on dictionaries and Internet resources.

Analysis and processing of the obtained data allows:

  • identify trends in the emotional background;
  • analyze student and teacher involvement in the learning process;
  • identify patterns of behavior;
  • carry out moderation of the teacher's work.

Recently, the burden on online schools has been growing greatly, the influx of students is becoming more and more. The number of lessons is increasing, it is increasingly difficult to monitor the quality of training. The system, developed by the ProgKids team, relieves the burden of methodologists and delegates control to neural networks. Artificial intelligence collects and analyzes, data allowing you to highlight problem areas and solve the shortcomings that have arisen locally, with the connection of school methodologists. The system is also being actively implemented in the field of private education. For example, in working with a tutor. At the end of each class, the AI sends the parent a brief information on how the lesson went.