Design and Implementation of Artificial Intelligence Online Learning Platform Based on Resource Scheduling Technology

Introduction

In today's large-scale artificial intelligence (AI) training, manual CPU resource allocation is inefficient due to task allocation and fault restart issues (Kayikci & Khoshgoftaar, 2024). To address these issues and improve the efficiency of artificial intelligence training, in this paper, we propose a novel machine learning platform with consistent server resource management and user specified resource allocation capabilities. Through the application of this machine learning platform, we can avoid problems such as low cluster resource utilization and uneven machine load distribution. The platform can automatically schedule and allocate AI tasks based on the current workload while utilizing containerization technology to build an efficient artificial intelligence system to handle these challenges. Automated training systems make the AI training process more convenient and efficient. One of the key functions of the system is to store a large amount of data, mirror and publish content on multiple servers, and automatically convert the trained model into highly available online API services. When a failure occurs, the system can also automatically restart to ensure the continuity and stability of the training task. The system we introduce in this article adopts a deep learning-based online learning (DLOL) platform, which combines the advantages of fuzzy convolutional neural networks (FCNN). To test the performance of the system, we adopted an improved blockchain gradient boosting decision tree (mGBDT) as the classification model. The classification model we propose in this article has multiple definitions and is suitable for different scenarios and needs. Through the machine learning platform and system we propose here, we can more efficiently conduct large-scale artificial intelligence training and achieve automated resource management and scheduling, providing strong support for the development of the AI field.

The contribution of the study are as follows:

• (1)

  The goal is to create an AI-powered online learning platform with built-in resource management and time-tracking capabilities.

• (2)

  The resources are automatically assigned by the OL platform using efficient cluster-based scheduling methods. These methods efficiently deal with both inefficient resource utilization and unequal load distribution.

• (3)

  To facilitate sequential model integration, the proposed model suggests a blockchain and deep learning-based OL platform (DLOL) that streamlines and fine-tunes the model's operational environment.

• (4)

  The OL platform is implemented using the decision tree algorithm and fuzzy-based convolutional neural networks.

Literature Review

AI education refers to the use of artificial intelligence in the education process (AIEd). Today, schools and universities all over the country are utilizing AIEd-based applications to various degrees (Ajani et al., 2024). Virtual machines capable of reasoning, problem solving, and learning will increase as a result of the development of efficient artificial intelligence programs. Using computers to process intelligence in the areas of language, mathematics, practicality, interpersonal, and intrapersonal is known as artificial intelligence (Kumar et al., 2024). Technologies that aid in the development of AI include natural language processing (NLP), expert systems, fuzzy logic, neural networks, and robotics, among others. Flight tracking programs and emergency care systems can both benefit from this combination of NLP and AI (Balobaid et al., 2024). Natural language processing provides an environmentally friendly language translation and development platform using virtual agents (AL-Akhras et al., 2024). Teachers and web-based engines work together to provide the best courses for students in the educational sector's most recent efforts. With the help of AI, educators can design classrooms that are free of prejudice. There has been an increase in global accessibility and interconnection and the interconnection of classrooms (Zhou et al., 2024). AI monitors learner mental actions such as self-regulation, control, and description, among others, in order to develop intelligent teacher programs. The students’ needs determine which information is most relevant. Their critical thinking and learning strategies can be aided by the use of AI.