A Knowledge Point Based Smart Classroom Teaching Assistant System

In recent years,individual learning path planning has prevailed in various online learning systems.The teaching planning modules in these online learning systems collect students' learning trajectories,and then help students find the most suitable learning plan or plan through their combination optimization or machine learning algorithm based on them,to achieve faster and better learning targets.Correspondingly,for teachers performing classroom teaching,little research has focused on how to teach to benefit more students.Meanwhile,the existing teaching assistant related work mostly focuses on the development of a single application,such as the roll call system and the homework system,etc.,and few have developed a comprehensive information system to assist the traditional courses throughout the entire teaching process.This thesis designs and implements a knowledge point based smart classroom teaching assistant system to obtain immediate and accurate students feedback,and further help teachers to understand the situation of students and quickly adjust teaching strategies to better achieve teaching goals.The system consists of the teacher's mobile phone,the student's smartphone/watch,the PC,and the background server.The main work of this thesis includes:(1)In conjunction with the actual situation in a traditional classroom,we analyze the functional requirements that an intelligent teaching assistant system based on knowledge points needs to meet.The purpose is to enable teachers to quantify,model,track,plan,and guide student learning on a quantitative and computable basis of knowledge points,as the first step toward observable,recordable,and computable classroom instruction;(2)Based on the system demand analysis,we research and implement the system's core algorithm.First,we propose a more interactive Bayesian Knowledge Tracking model.Then based on the established student model,we design and implement a genetic algorithm-based grouping recommendation algorithm and learning recommendation algorithm based on reinforcement learning for students.Finally,a teacher-oriented group teaching recommendation algorithm is proposed to provide teachers with better teaching arrangement recommendations for the entire class;(3)Then we implement the coding work according to the detailed design content of the system.The system is implemented based on a distributed architecture,which uses technologies such as distributed framework,cache database,relational database,frontend,back-end,Android development,and machine learning to ensure the realization of functions and business processes in system requirements analysis;(4)In the system development process,a complete set of testing procedures is designed to ensure the integrity and effectiveness of the system functionality,and at the same time to ensure the stability and reliability of the system in meeting the requirements analysis.