Статья описывает результаты анализа эффективности онлайн-курсов на платформе «Цифровой Наноград» с использованием алгоритмов машинного обучения. Исследование выявило, что малый объем курса и модульность оказывают наибольшее положительное влияние на эффективность курсов. Важную роль также играют интерактивные форматы представления информации и задания с автоматической проверкой. Модель, обученная с помощью алгоритма случайный лес, показала наилучшие результаты. В статье представлен анализ эффективности онлайн-курсов с использованием методов машинного обучения. Основное внимание уделено разработке и применению моделей, способных предсказать успех учащихся на основе различных факторов, таких как активность на платформе, взаимодействие с материалами курса и демографические данные. Проведено сравнение различных алгоритмов машинного обучения, включая логистическую регрессию, деревья решений и нейронные сети, для определения наиболее точного подхода. Результаты исследования демонстрируют значительное улучшение точности прогнозирования успеваемости, что позволяет образовательным платформам персонализировать обучение и повышать качество предоставляемых услуг.
The article describes the results of an analysis of the effectiveness of online courses on the «Digital Nanograd» platform using machine learning algorithms. The study found that course volume and modularity have a significant positive impact on course effectiveness. Interactive formats for presenting information and assignments with automatic checking also play a significant role. The model trained using the random forest algorithm showed the best results. The article describes the results of an analysis of the effectiveness of online courses on the "Digital Nanograd" platform using machine learning algorithms. The study found that a small course volume and modularity have the most positive impact on course effectiveness. Interactive formats of presenting information and assignments with automatic checking also play an important role. A model trained using the random forest algorithm showed the best results. The article presents an analysis of the effectiveness of online courses using machine learning methods. The main focus is on the development and application of models capable of predicting student success based on various factors such as platform activity, interaction with course materials, and demographic data. A comparison of various machine learning algorithms, including logistic regression, decision trees, and neural networks, was conducted to determine the most accurate approach. The research results demonstrate a significant improvement in the accuracy of performance prediction, allowing educational platforms to personalize learning and improve the quality of services provided.
The article describes the results of an analysis of the effectiveness of online courses on the «Digital Nanograd» platform using machine learning algorithms. The study found that course volume and modularity have a significant positive impact on course effectiveness. Interactive formats for presenting information and assignments with automatic checking also play a significant role. The model trained using the random forest algorithm showed the best results. The article describes the results of an analysis of the effectiveness of online courses on the "Digital Nanograd" platform using machine learning algorithms. The study found that a small course volume and modularity have the most positive impact on course effectiveness. Interactive formats of presenting information and assignments with automatic checking also play an important role. A model trained using the random forest algorithm showed the best results. The article presents an analysis of the effectiveness of online courses using machine learning methods. The main focus is on the development and application of models capable of predicting student success based on various factors such as platform activity, interaction with course materials, and demographic data. A comparison of various machine learning algorithms, including logistic regression, decision trees, and neural networks, was conducted to determine the most accurate approach. The research results demonstrate a significant improvement in the accuracy of performance prediction, allowing educational platforms to personalize learning and improve the quality of services provided.