Development and Application of an Online Autonomous Learning Platform Based on Artificial Intelligence

In response to the shortcomings of existing online self-learning platforms in personalized recommendations and real-time interactive support, this study aims to develop an online self-learning platform based on natural language processing (NLP) technology. Firstly, this article developed an intelligent question answering system using NLP technology, allowing students to interact with the platform through natural language. In addition, in order to more accurately adapt to the learning needs of different students, this article chooses the BERT (Transformer Bidirectional Encoder Representation) algorithm model to analyze students' learning behavior and historical data, in order to provide personalized suggestions for learning content and paths. Finally, by integrating an adaptive testing system based on NLP technology, the difficulty and type of questions can be dynamically adjusted according to students' real-time learning performance, achieving a highly personalized learning experience. The experimental results show that the average recommendation accuracy of this platform is 87.5<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">%</sup>, which is significantly higher than other learning platforms. The study also found that through learning on this platform, students' self-learning ability and problem-solving ability have been significantly improved.