Method, device and computer program for operating machine learning framework having active learning technique applied thereto

Abstract

The present invention relates to a user analysis method, carried out by a data analysis server, the method comprising: step A for configuring a question database including a plurality of questions, collecting solving result data, for the questions, of a user, and by using the solving result data, generating a data analysis model for modeling the user; step B for generating an expert model for suggesting data necessary for the machine learning of the data analysis model; step C for, according to a recommendation from the expert model, extracting at least one question from the question database and updating the data analysis model by using solving result data, for the extracted question, of the user; and step D for updating the expert model by applying, to update information of the data analysis model, a reward configured so as to improve the prediction accuracy of the data analysis model.

Description

technical field [0001] The present disclosure relates to a method of providing user-customized content using a data analysis framework. More specifically, the present disclosure relates to a method of generating a problem- and / or user-specific analysis model using a large amount of user content consumption result data and operating an expert model to select data required to effectively learn the analysis model. Background technique [0002] Typically, to date, educational content is provided in a packaged form. For example, a printed workbook with at least 700 questions, online or offline lectures containing study material that should be studied for at least a month, are sold together in one-hour or two-hour units. [0003] However, since all students have different individual weaknesses and weak question types, students need individually tailored content rather than packaged content. This is due to the fact that selectively studying only the weak question types in the wea...

AI Technical Summary

Problems solved by technology

[0004] However, students as learners themselves have difficulty identifying their own weaknesses

Also, in the traditional education field, since private educational institutions or publishing companies analyze students and questions based on subjective experience and intuition, it is not easy to provide questions optimized for individual students

[0005] As a result, in traditional educational environments, it is difficult for learners to provide personalized and customized content for more effective learning, and students may feel unfulfilled and quickly lose interest in packaged educational content

Images