Classification and analysis of MOOCs learner’s state: The study of hidden Markov model

Abstract

In MOOCs, learner's state is a key factor to learning effect. In order to study on learner's state and its change, the Hidden Markov Model was applied in our study, and some data of learner were analyzed, which includes MOOCs learner's basic information, learning behavior data, curriculum scores and data of participation in learning activities. The relationship of the learning state, the environment factors and the learner's individual conditions was found based on the data mining of the above of learning behavior data. Generally, there are three main conclusions in our research. Firstly, learners with different educational background have different learning states when they first learn from MOOCs. Secondly, the environmental factors such as curriculum quality, overall learning status and number of learners will influence the change of learners' learning status. Thirdly, the learner's behavioral expression is an observational signal of different learning states, which can be used to detect and manage the learner's learning states in different periods. From the analysis results of Hidden Markov Model, it is found that learners in different learning states can adopt appropriate methods to improve their learning efficiency. If the learner is in a negative state, the learning efficiency can be improved by improving the learning environment. If the learner is in a positive state, the positive learning status of the surrounding learners can help him or her maintain current state. Our research can help the MOOCs institutions improve the curriculum and provide reference for the development of MOOCs teaching.

Cultivation of Students’ Independent Learning Ability in Teaching Chinese as a Foreign Language Based on the CDIO Model of Guiding Data Structures and Hidden Markov Algorithms

“…In the specific experimental process, first, the data were transformed into a learning attitude dataset, learning initiative dataset, and learning effect dataset according to the definition of the observed dataset; then, the parameters of the student learning behavior model were obtained using the observed dataset in weeks 1-5 using the EM algorithm, and the obtained parametric model was applied to the observed dataset in weeks 6-9 for model validation. e model parameters and observed student data were then used to predict student learning behaviors over the next 10-15 weeks, and the weekly predictions were used to disrupt some students' learning [29,30].…”

Face recognition based on full convolutional neural network based on transfer learning model

“…Literature pointed out that the use of multi-scale extraction of face features can improve the accuracy of the algorithm to a certain extent, but it will significantly increase the amount of calculation and is not conducive to engineering implementation. Based on this, an adaptive scale feature extraction method is proposed to reduce feature dimensions and improve the computing speed [18,19]. Feature localization is the first step of feature extraction.…”

Quantifying the Impact of Students' Semester Course Load on Their Academic Performance

“…Results indicate that students with a full-time enrollment strategy have higher educational outcomes compared to students with mixed and part-time enrollment strategies. In another study, Chen et al [20] use HMM to categorize students based on their learning styles and investigate the relationship between learning style states and learning efficiency in massive open online courses (MOOCS).…”