With the rapid development of information technology and the widespread application of network technology, it has had a great impact on traditional learning methods. Teaching and learning based on the network environment has become an important form and part of school teaching. As a new type of learning method, online learning has its unique advantages and potential for development. It breaks through the limitations of time and space in traditional teaching methods, and brings great convenience and freedom to learners, but there are also many problem.
The purpose of this article is to analyze the effective learning behavior of students on the Internet based on data mining.This article analyzes learning behavior and builds a learning behavior mining model on this basis. Use the improved algorithm to mine and analyze the learning behavior data, including preprocessing and conversion of the data in the database, respectively mining the relationship between the learning sequence pattern and the learning behavior and effect, and finally the mining results explanation and evaluation. This paper does certain research and improvement on association rules and sequential pattern mining algorithms, and on this basis, uses the improved algorithm to collect, mine and analyze learning behavior data, and finally apply data mining technology to personalized the learning system optimizes the mechanism of online learning and the learning experience of users. Research on related methods of learning behavior mining, detailed analysis of association rules and sequence pattern mining, and optimization of Apriori algorithm and AprioriAll algorithm. The performance of the two algorithms and their optimized algorithms are compared respectively.
According to the data characteristics of the test data set, previous research uses the KNN, SVM, and RF algorithm to predict the learning state of students in the sample data set of the learning platform. The data set is randomly divided into 8:2 training data set and test data set, and the model training is conducted on the training data set, and the prediction test is conducted on the test data set. Then compared the predicted learning state results with the actual learning state results of the test data set, and the classification prediction accuracy of the three algorithms.
Above algorithms got less (up to 70%) performance accuracy, but they need more improvement.
To overcome those problems here we does certain research and improvement on association rules and sequential pattern mining algorithms, and on this basis, uses the improved algorithm to collect, mine and analyze learning behavior data, and finally apply data mining technology to personalized the learning system optimizes the mechanism of online learning and the learning experience of users. Research on related methods of learning behavior mining, detailed analysis of association rules and sequence pattern mining, and optimization of Apriori algorithm and Apriori All algorithm.
- The algorithm is exhaustive, so it finds all the rules with the specified support and confidence
- High performance
Algorithms: Apriori algorithm and Apriori All algorithm.
System : Pentium IV 2.4 GHz or intel
Hard Disk : 40 GB.
Floppy Drive : 1.44 Mb.
Mouse : Optical Mouse.
Ram : 512 Mb.
This article first introduces the meaning and background of the topic, and elaborates on the current situation of online learning behavior analysis at home and abroad and the development of data mining technology. This article describes the current development of online learning behavior and data mining technology at home and abroad. After the analysis, the research purpose of this article is put forward. This article introduces and analyzes the association rule algorithm andsequence pattern algorithm in data mining in detail, including related concepts and the process of classic algorithms. And based on the analysis of the original algorithm, a certain optimization is carried out. Finally, the performance of the two algorithms is compared and tested. This paper constructs a network learning behavior mining model and uses the improved association rule mining method to analyze the learning behavior, which mainly includes the preprocessing of the data in the data, the mining of the learning sequence pattern and the association between the learning behavior and the effect Mining relationships, and interpreting and evaluating the mining results.