Characteristics Of Learning Behavior Research Articles

Student's Behavior Clustering based on Ubiquitous Learning Log Data using Unsupervised Machine Learning

Online learning is the source of data generation related to learner's learning behaviors, which is valuable for knowledge discovery. Existing research emphasized more on an understanding of student's performance and achievement from learning log data. In this study, we presented data-driven learning behavior clustering in authentic learning context to understand students' behavior while participating in the learning process. The objective of the study is to distinguish students according to their learning behavior characteristics and identify clusters of students at risk of unsuccessful learning achievement. Learning log data were collected from ubiquitous learning applications before conducting Exploratory Data Analysis (EDA) and cluster analysis. We used partitional clustering using K-means algorithm and hierarchical clustering based on the agglomerative method to improve clustering strategies. The result of this study revealed three different clusters of students supported by data visualization techniques. Cluster 1 comprised more students with active learning behavior based on the total logs, total problems posed, and the total attempts in fraction operation and simplification. Students in clusters 2 and 3 had a higher attempt at problem-solving instead of problem-posing. Both clusters also focused on fraction's conceptual understanding. Knowledge discovery of this study used real data generated from ubiquitous learning application namely U-Fraction. We combined two different types of clustering method for delivering more accurate portrait of a student's hidden learning behaviors. The outcome of this study can be a basis for educational stakeholders to provide preventive learning strategies tailored to a different cluster of students.

Personality and Learning Behavioral Characteristics as Predictors of Academic Achievement of Medical Students

This study investigates whether personality characteristics and learning behaviors can predict medical students’ academic achievement in Korea, specifically in terms of successfully completing medical school without delays or achieving a high grade point average (GPA) in their final year. In May 2018, 316 medical students took the Multi-Dimensional Learning Strategy Test, 2nd edition, which provided data on their personality and learning behavioral characteristics. Their final year’s GPA and any delays in completing medical school were ascertained by reviewing all electronic academic records of each semester they had been enrolled. The combination of personality and learning behavioral characteristics was significantly associated with completing medical school without delays, even after adjusting for sex and admission path. A multiple logistic regression analysis showed that the adjusted odds ratios and 95% confidence intervals for completing medical school without delays were 1.52 (95% confidence interval [CI], 0.83–2.78) and 3.64 (95% CI, 1.70–7.82) for “others” and “both high” categories, respectively, when compared with the “both low” category. For 235 students who completed medical school without delays, their learning behavioral characteristics (scores) were significantly associated with their final year’s GPA even after adjusting for sex, admission path, and personality characteristics (scores) as determined by the multiple linear regression analysis. This study suggests that individual personality and learning behavior characteristics are predictors of medical students’ academic achievement. Therefore, interventions such as personalized counseling programs should be provided in consideration of such student characteristics.

Research on the Mechanism of Dynamic Monitoring of Undergraduate Students’ Learning Situation and Adaptive Adjustment of Teaching Mode--Based on the Perspective of Educational Evaluation Reform

Abstract Dynamic monitoring of undergraduate learning is a tool for the accurate pulse of education and teaching, and an in-depth study of the quality monitoring data of students’ learning can discover the hidden problems and shortboards in education and teaching. The article utilizes information technology to build a dynamic monitoring platform for undergraduates. Then, it establishes an accurate teaching model for undergraduates with the assistance of this platform. The SHAP interpretable model is used to obtain the online learning behavior characteristics of undergraduates, and the LSTM autocoder is used to construct the time-varying feature sequence of undergraduates’ learning behavior, which is inputted into the LSTM model to establish the undergraduate learning situation early warning model. Taking the data of students’ online behavioral characteristics as an example, the validation of the importance of undergraduate learning behavioral characteristics and early warning is carried out, and the effect of the precise teaching model is also analyzed. The SHAP value of undergraduate students taking online tests is 0.969, which significantly impacts their online learning behavior. The accuracy of undergraduate students’ learning alert was 0.822, which was about 3.53% higher than the FWTS-CNN model with sub-optimal performance, and the results of the learning adaptability retest were 3.24 points higher than the initial test results. From the perspective of educational evaluation reform, combining undergraduate students’ learning dynamic monitoring data can enable adaptive adjustment of teaching content and mode.

Application of mathematical modeling in analyzing and optimizing English teaching methods in vocational education

Abstract In the era of big data, mathematical and scientific methods significantly contribute to educational reform, continuously revitalizing precision teaching. In this paper, we use the new Hawkes process to collect and vectorize data on how students learn. Then, we use the unique thermal coding method to combine the behavioral feature vector with the teaching feature vector. Finally, we classify the data and use the softmax function to predict the learning effect. Based on the prediction results, a mathematical model for precise teaching optimization of vocational education English has been established, and the decision function is used to solve the problem and optimize teaching methods. It was found that all the indicators of learning behavior characteristics of students in the experimental class applying the teaching method optimization model were significantly improved, and there was a significant difference at the level of 0.05 (P = 0.000<0.05). Furthermore, the students in the experimental class achieved higher academic performance in vocational education English than those in the control class under traditional teaching. This paper provides a scientific basis for teaching English to students in vocational education and provides methodological references and ideas for unfolding individualized and precise learning in classroom teaching.

Mining method of students’ learning behavior characteristics in online classroom of Colleges and Universities based on dense clustering method

Mining method of students’ learning behavior characteristics in online classroom of Colleges and Universities based on dense clustering method

Analysis of the Design and Effectiveness Evaluation of Personalized English Teaching System for Mongolian College Students in Trilingual Context

Abstract In order to promote the efficient development of English teaching in Mongolian colleges and universities, this paper constructs an English-personalized teaching system based on trilingual background. Firstly, BP neural network algorithm is used to manage and classify students’ learning data information and detect system teaching evaluation indexes by identifying students’ learning behavior characteristics. Secondly, the K-mean clustering algorithm is invoked in the detection process to improve the detection efficiency to ensure the accuracy of obtaining teaching evaluation indexes. Finally, to verify the effectiveness of the teaching system, a practical analysis of its teaching effect is conducted. The analysis results show that the completion rates of the teaching system constructed in this paper are 78.7% and 86% in course learning and course testing, respectively, and 81% and 75% in information management and knowledge management, respectively. And in the experimental group, the grade passing rate was generally improved by about 42.3%. It proves that the personalized English teaching system constructed in this paper effectively improves the teaching quality of Mongolian college students’ English and is beneficial to the long-term development of English teaching in Mongolian colleges and universities.

Study of online learning time and learning performance model based on learning platform

Abstract In the new online learning environment where information technology and education are integrated, it is important to understand the learning patterns of online learners and then predict their learning outcomes to improve learning effectiveness of online learners. In this paper, we collect students’ online learning behavior data through an online learning platform and further investigate the influence mechanism between online learning behavior characteristics and learning performance with the help of the OCCP classification model and the 3D S-F-T model. Meanwhile, the Light GBM algorithm is based on its advantages in noise and distributed processing based on its high accuracy and strong generalization ability in classification problems. The Light GBM algorithm constructs the performance prediction model for the selected learning behavior feature data. The results show that the accuracy of academic achievement prediction based on the Light GBM model is 0.85, and the accuracy of sample prediction classification is H (85%), M (73%) and L (89%), indicating that the model is effective for academic achievement prediction. The experimental results confirm that the prediction method can more accurately identify at-risk students and provide more intuitive and accurate feedback on learners’ problems and deficiencies in the learning process to support the implementation of personalized education interventions.

MOOC performance prediction and analysis via Bayesian network and Maslow’s hierarchical needs theory

ABSTRACT Nowadays, Massive Open Online Courses (MOOC) has been gradually accepted by the public as a new type of education and teaching method. However, due to the lack of timely intervention and guidance from educators, learners’ performance is not as effective as it could be. To address this problem, predicting MOOC learners’ performance and providing them with timely interventions have become an indispensable part for the MOOC learning. However, current MOOC performance prediction methods cannot provide us with interpretable prediction results and cannot further help us to provide learners with targeted intervention strategies. To this end, we adopt the framework of Bayesian Network (BN) and then constructed an MOOC Performance Prediction BN (MPBN), which provides us with a graphical explanation of how learners’ demographical and learning behavior characteristics affect their performance. Besides, since the productive MOOC learners tend to be driven by their inner goals, we further use Maslow’s hierarchical needs theory to construct several indicators, by which to analyze the prediction of MPBN and then propose the appropriate intervention strategies.

An Analysis of Online Learner Types Applicable to Lifelong Learning Environments

<p>Based on the perspective of Sustainable Development Goals (SDGs) Quality Education and lifelong learning, it is necessary to respect the learning opportunities and quality for all individuals. Online learning can provide more opportunities for lifelong learning, but due to the significant differences in students’ backgrounds and characteristics, personalized and timely support becomes more crucial. Learning analytics (LA) in online learning environment is a way to facilitate understanding of the potential meaningful information and relationships of students. One of the main functions of LA is to monitor the learning performance and identify potential learning problems early. In this study, 𝑘-means clustering is performed to determine the types of learning in lifelong online learning environments, based on students’ personal traits (background factors), learning behavior paths, and interactive perspectives on learning performance. Moreover, statistical analysis is used to further evaluate the linear correlation coefficients as well as the characteristics of each group of students, who ranged in age from 18 to 73, with a total of 2386 participants from five courses, in the interactive perspective. The result shows a significant correlation between learning performance and persistence across the three learning clusters, with a tendency towards continuous learning, thus providing educators an understanding of the learning behavior characteristics of those types of online learners.</p> <p> </p>

Student Online Learning Behavior Characteristics Based on Multidimensional Cognitive Model

Analysis of student learning behavior characteristics is an important means for educators to better understand students and improve the quality and effectiveness of teaching in the field of education. It is necessary to refer to students' cognitive levels for analysis of student learning behavior characteristics. However, existing algorithms only focus on the overall performance and grades of students, ignoring the individual differences in learning cognitive levels among students, which affects the accuracy of the analysis results. Therefore, this paper conducts research on student online learning behavior characteristics based on a multidimensional cognitive model. Firstly, a multidimensional and multilevel model for evaluating students' cognitive levels is constructed, and the process of evaluating students' cognitive levels is sustainable and can be adjusted in real-time as students' cognitive levels change. By considering the differences in evaluation levels and students' cognitive levels, targeted observation and extraction of students' online learning behavior characteristics can be achieved. A new model based on variational autoencoder neural network is proposed to perform decoupled representation of students' implicit preferences. By using a regularization term based on maximum mean difference, the model can learn independent hidden vectors sensitive to dynamic and static factors from students' online learning behavior history data and multidimensional cognitive evaluation history data. The experimental results verify the effectiveness of the constructed model.

Prediction of student online Learning performance based on Machinelearning

This essay uses machine learning methods to predict students’ online learning performance. The characteristicsof student’s characteristics, learning behavior characteristics, and environmental factors are extracted by featureengineering, and the random forest model is used for prediction. The experimental results show that the predictionaccuracy of the random forest model reaches 0.82, which provides an effective means for the prediction of onlinelearning performance.

Analysis of Learning Behavior Characteristics and Prediction of Learning Effect for Improving College Students’ Information Literacy Based on Machine Learning

Information literacy is a basic ability for college students to adapt to social needs at present, and it is also a necessary quality for self-learning and lifelong learning. It is an effective way to reveal the information literacy teaching mechanism to use the rich and diverse information literacy learning behavior characteristics to carry out the learning effect prediction analysis. This paper analyzes the characteristics of college students’ learning behaviors and explores the predictive learning effect by constructing a predictive model of learning effect based on information literacy learning behavior characteristics. The experiment used 320 college students’ information literacy learning data from Chinese university. Pearson algorithm is used to analyze the learning behavior characteristics of college students’ information literacy, revealing that there is a significant correlation between the characteristics of information thinking and learning effect. The supervised classification algorithms such as Decision Tree, KNN, Naive Bayes, Neural Net and Random Forest are used to classify and predict the learning effect of college students’ information literacy. It is determined that the Random Forest prediction model has the best performance in the classification prediction of learning effect. The value of Accuracy is 92.50%, Precision is 84.56%, Recall is 94.81%, F1-Score is 89.39%, and Kapaa coefficient is 0.859. This paper puts forward differentiated intervention suggestions and management decision-making reference in the information literacy teaching process of college students, with a view to adjusting the information literacy teaching behavior, improving the information literacy teaching quality, optimizing educational decision-making, and promoting the sustainable development of high-quality and innovative talents in the information society.Our work involving research of the thinking and direction of the sustainable development of information literacy training proved to be encouraging.

On the Cognitive Load of Online Learners With Multi-Level Data Mining

A lot of studies have shown that there is an “inverse U-curve” relationship between learners' grades and cognitive load. Learners' grades are closely related to their learning behavior characteristics on online learning. Is there any relationship between online learners' behavior characteristics and cognitive load? Based on this, the data of research are obtained from the professions and applied sciences on the Canvas Network platform. The multi-level data mining technology is used to analyze and mine the relationship between grades and online learners' behavior characteristics layer by layer. The results show that there is an “inverse U-curve” relationship between grades and “nevents.” Therefore, the research attempts to map “nevents” to the online learners' cognitive load, which makes the online learners' cognitive load can be quantitative analysis. Research results also prove that multi-level data mining technology can be used to mine the special learning rules hidden behind the data effectively.

Improving Random Forest Algorithm for University Academic Affairs Management System Platform Construction

Using the data of students’ learning process recorded in the network teaching platform to predict students’ learning performance, assist teachers to analyze learning situation, formulate teaching strategies, and warn about students’ learning state is a hot spot in the field of mixed curriculum research in recent years. In view of the complexity, heterogeneity, and security of college educational administration data and the difficulty of predicting and analyzing college students’ achievements, this paper designs a college educational administration management system platform based on improved random forest algorithm. Combining the advantages of three data-driven prediction algorithms, namely, random forest, extreme gradient boosting (XGBoost), and gradient boosting decision tree (GBDT), a model based on improved random forest algorithm is proposed. It is proved that this method is a noninferior prediction method. Secondly, the model is applied to practical problems to solve the problem of predicting college students’ grades. An experiment is carried out on the real data set provided by a municipal education bureau. The results show that the proposed model not only achieves good prediction accuracy, but also solves the stability problem of the model after adding new data, which will contribute to the iterative optimization of the model, improve the universality of the model, and help continuously track the learning behavior characteristics of college students in different semesters.

Dropout Rate Prediction of Massive Open Online Courses Based on Convolutional Neural Networks and Long Short-Term Memory Network

Massive open online courses (MOOC) is characterized by large scale, openness, autonomy, and personalization, attracting increasingly students to participate in learning and gaining recognition from more and more people. This paper proposes a network model based on convolutional neural networks and long short-term memory network (CNN-LSTM) for MOOC dropout prediction task. The model selects 43-dimensional behavioral features as input from students’ learning activity logs and adopts the CNN model to automatically extract continuous features over a period of time from students’ learning activity logs. At the same time, considering the time sequence of students’ learning behavior characteristics, a MOOC dropout prediction model was established by using long short-term memory network to obtain students’ learning status at different time steps. The algorithm proposed in this chapter was trained and evaluated on the public dataset provided by the KDD Cup 2015 competition. Compared with the dropout prediction methods based on LSTM and CNN-RNN, the model improved the AUC by 2.7% and 1.4%, respectively. The result in this paper is a good predictor of dropout rates and is expected to provide teaching aid to teachers.

Research on E-learning Behavior Evaluation of Students Based on Three-way Decisions Classification Algorithm

With the popularization and application of E-learning platform, it is feasible to collect learning behavior data of students, which provides data basis for analyzing the knowledge and rules contained in learning process of students. At present, the common e-learning platform only realizes the simple statistics and display function for the data, and does not do further in-depth calculation. The data observed by educators are still superficial learning phenomena, and they cannot see the learning rules behind the appearance. It is difficult to effectively guide students, change their learning routes and feedback teaching strategies. In view of this, the main characteristics of learning behavior deeply mined from learning activities and learning evaluation, and the important data feature items affect E-learning effect obtained. Three decision classification algorithms applied to establish prediction and evaluation model for students with learning behavior risk. The overall accuracy of model test is in line with the expectation. This provides a practical application modeling method for further implementation of personalized learning service.

Research on learning initiative Based on behavior quantization and potential value clustering

As an important learning mode, e-learning has become an important way of online learning, especially during the coronavirus pandemic. However, the separation of teaching and learning in time or space undoubtedly makes it difficult for teachers to master and supervise students' learning initiative. From the perspective of inquiry learning initiative, a learning problem model is constructed. To solve the problem of model timeliness, the learning behavior characteristics have been quantified with the guidance of time parameter. Furthermore, a cluster analysis method is proposed to group the students into different clusters, and a case study is given to verify the proposed scheme. The experimentalresults show that the proposed method can effectively identify the type of students and the degree of learning engagement.

Assessment and Relationships of Conditional Motivational Regulation Strategy Knowledge as an Aspect of Undergraduates' Self-regulated Learning Competencies

Abstract. For learners, motivational regulation represents a daily challenge in terms of initiating, increasing and maintaining motivation. Crucial here is conditional knowledge about which motivational regulation strategies are appropriate for which specific motivational problems. This paper addresses the relationships between this type of strategy knowledge and regulatory effectiveness and characteristics of learning behavior. In a sample of 188 undergraduates, this strategy knowledge – assessed with a newly developed Situational Judgement Test – showed moderately positive correlations with motivational regulation on the behavioral level. Moreover, it was positively related to the effectiveness of motivational regulation and effort. Hence, conditional motivational regulation strategy knowledge is relevant and contributes to an extended understanding of motivational regulation in the higher education context.

A multi-constraint learning path recommendation algorithm based on knowledge map

It is difficult for e-learners to make decisions on how to learn when they are facing with a large amount of learning resources, especially when they have to balance available limited learning time and multiple learning objectives in various learning scenarios. This research presented in this paper addresses this challenge by proposing a new multi-constraint learning path recommendation algorithm based on knowledge map. The main contributions of the paper are as follows. Firstly, two hypotheses on e-learners’ different learning path preferences for four different learning scenarios (initial learning, usual review, pre-exam learning and pre-exam review) are verified through questionnaire-based statistical analysis. Secondly, according to learning behavior characteristics of four types of the learning scenarios, a multi-constraint learning path recommendation model is proposed, in which the variables and their weighted coefficients considers different learning path preferences of the learners in different learning scenarios as well as learning resource organization and fragmented time. Thirdly, based on the proposed model and knowledge map, the design and implementation of a multi-constraint learning path recommendation algorithm is described. Finally, it is shown that the questionnaire results from over 110 e-learners verify the effectiveness of the proposed algorithm and show the similarity between the learners’ self-organized learning paths and the recommended learning paths.

Individual variability in visual discrimination and reversal learning performance in common marmosets

Detailed information about the characteristics of learning behavior in marmosets is useful for future marmoset research. We trained 42 marmosets in visual discrimination and reversal learning. All marmosets could learn visual discrimination, and all but one could complete reversal learning, though some marmosets failed to touch the visual stimuli and were screened out. In 87% of measurements, the final percentage of correct responses was over 95%. We quantified performance with two measures: onset trial and dynamic interval. Onset trial represents the number of trials that elapsed before the marmoset started to learn. Dynamic interval represents the number of trials from the start before reaching the final percentage of correct responses. Both measures decreased drastically as a result of the formation of discrimination learning sets. In reversal learning, both measures worsened, but the effect on onset trial was far greater. The effects of age and sex were not significant as far as we used adolescent or young adult marmosets. Unexpectedly, experimental circumstance (in the colony or isolator) had only a subtle effect on performance. However, we found that marmosets from different families exhibited different learning process characteristics, suggesting some family effect on learning.