Small-scale Online Teaching Platform Development And Research On Recommendation Of Exercises

With the rapid rise of mobile Internet,the in-depth development of network services and the high popularity of smartphones,online teaching has become an inevitable trend.Various online teaching platforms have sprung up,but many applications still use the traditional monolithic architecture model.Due to the continuous growth of users and the increasing demand of users,the traditional monolithic architecture model has been unable to adapt to the rapidly developing Internet era.The industry's scalable system architecture complements the shortcomings of traditional monolithic architectures.This paper draws on the concept of AKF scalable cubic and micro-services,and proposes a small-scale online teaching platform development model.The model uses the Fourier series to orthogonally decompose the function,and maps the programming space of the traditional online teaching platform to the multi-dimensional orthogonal space according to the unrelated attribute set,and then combines the space into multi-dimensional orthogonal dimension.A functional module with different granularity,and then import the module code into the virtual machine of the cloud platform.Through this method of decomposition mapping and virtualization,the development logic of the online teaching platform is easier to understand,the complexity is lower,the code amount is smaller,and the reuse rate is higher.When a large-scale user accesses the system at the same time,the server needs to process a large number of access requests in a short period of time.With such high concurrency,the system response time will become longer or even prone to crashes.This paper proposes an online test system load balancing method based on learning analysis and prediction model for the problem of load imbalance caused by high concurrent interviews in a large number of students participating in the online test system.The method firstly analyzes the relationship between the number of concurrent connections between the database and the user's answering behavior according to the user's online test data,and reduces the number of concurrent connections by reasonably controlling the test interval.Then,the students' total number of questions,the number of collection questions,and the average answer time are extracted.The average time of the test is used as a label,and the candidate's answer time prediction model is constructed by convolutional neural network.Finally,all students are classified and redirected to different servers based on the average length of the exam.By comparing the experimental results,the method can effectively reduce the number of concurrent connections in the database,alleviate the problem of high concurrent access leading to system instability,and make the load of each server more balanced.And the method can predict the load situation in advance,and provides a reference for purchasing server resources before the test.learners are increasingly demanding the personalization and practicality of exercise.Facing the massive exercises in the online learning platform and the online examination system,how to choose the exercises that can be targeted and can make up for the knowledge loopholes has become a hot topic in the current field of personalized recommendation of teaching resources.In view of the fact that learners have a variety of learning features,and there are a large number of online exercises,various types and varying degrees of difficulty,this paper proposes a precise recommendation method based on Multidimensional features analysis for exercises.It quantifies the potential relationship between the learner and the exercise from three aspects:the heat of the exercise itself,the relevance of the knowledge among the exercises,and the similarity of the learner's style,using the linear combination and the learning to rank method to build the recommendation model to match learner and exercises exactly.Experiments show that the mean average precision of this method reaches 36.8%when recommending five candidate exercises,which provides learners with personalized exercise recommendation services and ultimately improves the learning efficiency.