Analysis and Prediction of Student Performance Based on Moodle Log Data using Machine Learning Techniques

During the COVID-19 Pandemic, many universities in Thailand were mostly locked down and classrooms were also transformed into a fully online format. It was challenging for teachers to manage online learning and especially to track student behavior since the teacher could not observe and notify students. To alleviate this problem, one solution that has become increasingly important is the prediction of student performance based on their log data. This study, therefore, aims to analyze student behavior data by applying Predictive Analytics through Moodle Log for approximately 54,803 events. Six Machine Learning Classifiers (Neural Network, Random Forest, Decision Tree, Logistic Regression, Linear Regression, and Support Vector Machine) were applied to predict student performance. Further, we attained a comparison of the effectiveness of early prediction for four stages at 25%, 50%, 75%, and 100% of the course. The prediction models could guide future studies, motivate self-preparation and reduce dropout rates. In the experiment, the model with 5-fold cross-validation was evaluated. Results indicated that the Decision Tree performed best at 81.10% upon course completion. Meanwhile, the SVM had the best result at 86.90% at the first stage, at 25% of the course, and Linear Regression performed with the best efficiency at the middle stages at 70.80%, and 80.20% respectively. The results could be applied to other courses and on a larger e-learning systems log that has similar student activity conditions and this could contribute to more accurate student performance prediction © 2023, International Journal of Emerging Technologies in Learning.All Rights Reserved.

PREDICTIVE ABILITY OF FINANCIAL AND NON-FINANCIAL PERFORMANCE FOR FINANCIAL STATEMENT PUBLICATION TIME FRAME: MODERATING ROLE OF COVID-19 PANDEMIC

Objetivo: Este estudo examinou a capacidade de desempenho financeiro e nao financeiro na previsäo do tempo de publicaçao de relatórios financeiros, moderada pela pandemia da COVID-19. Referenciái teórico: A teoria dos sinais postula que a administraçâo desempenha um papel crucial no fornecimento de informaçöes as partes interessadas sobre as condiçöes da empresa (Brigham & Houston, 2001). De acordo com Spence (1973), as empresas estao motivadas a fornecer informaçöes relevantes as partes interessadas. Se as condiçöes de desempenho sao boas, a empresa tende a acelerar o processo de apresentaçao de demonstraçöes financeiras. Por outro lado, se o desempenho for ruim, há uma tendencia a atrasar a publicaçao dos relatórios financeiros. O longo período de tempo para a publicaçao de relatórios financeiros pode indicar más noticias que a empresa tem, de modo que ela ainda tem que publicar as noticias para o público. Scott (2015) sugere que quando os gerentes souberem que há noticias desfavoráveis sobre a condiçao da empresa no futuro, evitarao publicar estas informaçöes ou pelo menos atrasaräo a apresentaçao das demonstraçöes financeiras. Método: O desempenho financeiro foi medido por quatro indicadores: lucratividade, liquidez e solvencia. Enquanto isso, o desempenho nao financeiro variável foi medido pelo indice de boa governança corporativa (GCG) e pela reputaçao dos auditores. O modelo proposto foi testado com base nos dados quantitativos coletados de 156 empresas de manufatura listadas na Bolsa de Valores da Indonesia (IDX) a partir de 2018 e 2020. A análise de regressao múltipla foi realizada para analisar e interpretar os dados. Resultados e conclusao: O resultado indica que a solvencia, a boa governança corporativa e a reputaçao do auditor foram preditores significativos do período de publicaçao do relatório financeiro. Entretanto, a capacidade preditiva de rentabilidade e liquidez no prazo de publicaçao nao foi considerada significativa. Além disso, os resultados mostram que a pandemia da COVID-19 modera a capacidade de rentabilidade e boa governança corporativa na previsao do prazo de publicaçao. Implicates da pesquisa: O indicador de desempenho financeiro e nao financeiro dá resultados diferentes na previsäo do RWPLK das empresas de manufatura na Indonesia. ROA e CR nao sao capazes de prever o RWPLK, mas DER, GCG, KAP sao capazes de prever o RWPLK. O papel da pandemia COVID-19 foi capaz de moderar a capacidade de ROA e GCG em prever o prazo para publicaçao de relatórios financeiros, mas foi incapaz de moderar a capacidade de CR, DER e KAP em prever o RWPLK. Originalidade/valor: O presente estudo fornece a primeira evidencia empírica sobre o papel moderador da pandemia COVID-19 na capacidade preditiva do desempenho financeiro e nao financeiro para o prazo de publicaçao das demonstraçöes financeiras.Alternate :Purpose: This study examined the ability of financial and non-financial performance in predicting financial reports publication time frame as moderated by the COVID-19 pandemic. Theoretical framework: Signal theory postulates that management serves a crucial role in providing information to stakeholders regarding the condition of the company (Brigham & Houston, 2001). According to Spence (1973), companies are motivated to provide relevant information to stakeholders. If the performance conditions are good, the company tend to speed up the process of presenting financial statements. Conversely, if performance is poor, there is a tendency to delay the financial reports publication. The long span of time for the publication of financial reports can indicate bad news that the company has so that it has yet to publish the news to the public. Scott (2015) suggests that when managers know there is unfavorable news about the condition of the company in the future, they will avoid publishing this information or at least delay the presentation of financial statements. Method/design/approach: Financial performance was measured by four indicators: profita il ty, liquidity and solvency. Meanwhile, variable non-financial performance was measured by the index of good corporate governance (GCG) and auditor reputation. The proposed model was tested based on the quantitative data collected from 156 manufacturing companies listed on the Indonesia Stock Exchange (IDX) from 2018 and 2020. The multiple regression analysis was performed to analyze and interpret the data. Results and conclusion: Result indicates that solvency, good corporate governance, and auditor reputation were significant predictors of the time span of financial report publication. However, the predictive ability of profitability and liquidity on the publication timeframe was found to be not significant. Furthermore, the results show that the COVID-19 pandemic moderates the ability of profitability and good corporate governance in predicting the publication timeframe. Research implications: Financial and non-financial performance indicator gives different results in predicting the RWPLK of manufacturing companies in Indonesia. ROA and CR are not able to predict RWPLK, but DER, GCG, KAP are able to predict RWPLK. The role of the COVID-19 pandemic was able to moderate the ability of ROA and GCG in predicting the timeframe for publication of financial reports, but was unable to moderate the ability of CR, DER and KAP in predicting RWPLK. Originality/value: The present study provides the first empirical evidence on the moderating role of the COVID19 pandemic on the predictive ability of financial and non-financial performance for financial statement publication time frame.

E-learning enhancement through educational data mining with Covid-19 outbreak period in backdrop: A review.

E-learning is fast becoming an integral part of the teaching- learning process, particularly after the outbreak of Covid-19 pandemic. Educational institutions across the globe are striving to enhance their e-learning instructional mechanism in accordance with the aspirations of present-day students who are widely using numerous technological tools - computers, tablets, mobiles, and Internet for educational purposes. In the wake of the evident incorporation of e-learning into the educational process, research related to the application of Educational Data Mining (EDM) techniques for enhancing e-learning systems has gained significance in recent times. The various data mining techniques applied by researchers to study hidden trends or patterns in educational data can provide valuable insights for educational institutions in terms of making the learning process adaptive to student needs. The insights can help the institutions achieve their ultimate goal of improving student academic performance in technology-assisted learning systems of the modern world. This review paper aims to comprehend EDM's role in enhancing e-learning environments with reference to commonly-used techniques, along with student performance prediction, the impact of Covid-19 pandemic on e-learning and priority e-learning focus areas in the future.

A Comprehensive Study to Investigate Student Performance in Online Education during Covid-19

During the recent Covid-19 pandemic, there has been a tremendous increase in online-based learning (e-learning) activities as nearly every educational institution has transferred its programs to digital platforms. This makes it crucial to investigate student performance under this new mode of delivery. This research conducts a comparison among the traditional educational data mining techniques to detect the best performing classifier for analyzing as well as predicting students' performance in online learning platforms during the pandemic. It is achieved through extracting four datasets from X-University student information system and learning platform, followed by the application of 6 classifiers to the extracted datasets. Random Forest Classifier has demonstrated the highest accuracy in the first two out of the four datasets, while Simple Cart and Naïve Bayes Classifiers presented the same for the remainder two. All the classifiers have demonstrated medium to high TP rates, class precision and recall, ranging from 60% to 100% for almost all of the classes. This study emphasized the attributes that have a direct impact on students' performance. The outcomes of this study will assist the instructors and educational institutions to identify important factors in the analysis and prediction of student performance for online program delivery.

Evaluation of Synthetic Categorical Data Generation Techniques for Predicting Cardiovascular Diseases and Post-Hoc Interpretability of the Risk Factors

Machine Learning (ML) methods have become important for enhancing the performance of decision-support predictive models. However, class imbalance is one of the main challenges for developing ML models, because it may bias the learning process and the model generalization ability. In this paper, we consider oversampling methods for generating synthetic categorical clinical data aiming to improve the predictive performance in ML models, and the identification of risk factors for cardiovascular diseases (CVDs). We performed a comparative study of several categorical synthetic data generation methods, including Synthetic Minority Oversampling Technique Nominal (SMOTEN), Tabular Variational Autoencoder (TVAE) and Conditional Tabular Generative Adversarial Networks (CTGANs). Then, we assessed the impact of combining oversampling strategies and linear and nonlinear supervised ML methods. Lastly, we conducted a post-hoc model interpretability based on the importance of the risk factors. Experimental results show the potential of GAN-based models for generating high-quality categorical synthetic data, yielding probability mass functions that are very close to those provided by real data, maintaining relevant insights, and contributing to increasing the predictive performance. The GAN-based model and a linear classifier outperform other oversampling techniques, improving the area under the curve by 2%. These results demonstrate the capability of synthetic data to help with both determining risk factors and building models for CVD prediction.

High-performance association rule mining: Mortality prediction model for cardiovascular patients with COVID-19 patterns

On a global scale, 213 countries and territories have been affected by the coronavirus outbreak. According to researchers, underlying co-morbidity, which includes conditions like diabetes, hypertension, cancer, cardiovascular disease, and chronic respiratory disease, impacts mortality. The current situation requires for immediate delivery of solutions. The diagnosis should therefore be more accurate. Therefore, it's essential to determine each person's level of risk in order to prioritise testing for those who are subject to greater risk. The COVID-19 pandemic's onset and the cases of COVID-19 patients who have cardiovascular illness require specific handling. The paper focuses on defining the symptom rule for COVID-19 sickness in cardiovascular patients. The patient's chronic condition was taken into account while classifying the symptoms and determining the likelihood of fatality. The study found that a large proportion of people with fever, sore throats, and coughs have a history of stroke, high cholesterol, diabetes, and obesity. Patients with stroke were more likely to experience chest discomfort, hypertension, diabetes, and obesity. Additionally, the strategy scales well for large datasets and the computing time required for the entire rule extraction procedure is faster than the existing state-of-the-art method. © 2023 Faculty of Engineering, Alexandria University

A Matrix Factorization-Based Drug-Virus Link Prediction Method for SARS-CoV-2 Drug Prioritization

Matrix factorization (MF) has been widely used in drug discovery for link prediction, which aims to reveal new drug-target links by integrating drug-drug and target-target similarity information with a drug-target interaction matrix. The MF method is based on the assumption that similar drugs share similar targets and vice versa. However, one major disadvantage is that only one similarity metric is used in MF models, which is not enough to represent the similarity between drugs or targets. In this work, we develop a similarity fusion enhanced MF model to incorporate different types of similarity for novel drug-target link prediction. We apply the proposed model on a drug-virus association dataset for anti-COVID drug prioritization, and compare the performance with other existing MF models developed for COVID. The results show that the similarity fusion method can provide more useful information for drug-drug and virus-virus similarity and hence improve the performance of MF models. The top 10 drugs as prioritized by our model are provided, together with supporting evidence from literature. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

Predicting User Satisfaction of Mobile Healthcare Services Using Machine Learning: Confronting the COVID-19 Pandemic

The outbreak of COVID-19 led to rapid development of the mobile healthcare services. Given that user satisfaction is of great significance in inducing marketing success in competition markets, this research explores and predicts user satisfaction with mobile healthcare services. Specifically, the current research aimed to design a machine learning model that predicts user satisfaction with healthcare services using big data from Google Play Store reviews and satisfaction ratings. By dealing with the sentimental features in online reviews with five classifiers, the authors find that logistic regression with term frequency-inverse document frequency (TF-IDF) and XGBoost with bag of words (BoW) have superior performances in predicting user satisfaction for healthcare services. Based on these results, the authors conclude that such user-generated texts as online reviews can be used to predict user satisfaction, and logistic regression with TF-IDF and XGBoost with BoW can be prioritized for developing online review analysis platforms for healthcare service providers.

Effects of Alcohol-Blended Waste Plastic Oil on Engine Performance Characteristics and Emissions of a Diesel Engine

The current study aims to investigate and compare the effects of waste plastic oil blended with n-butanol on the characteristics of diesel engines and exhaust gas emissions. Waste plastic oil produced by the pyrolysis process was blended with n-butanol at 5%, 10%, and 15% by volume. Experiments were conducted on a four-stroke, four-cylinder, water-cooled, direct injection diesel engine with a variation of five engine loads, while the engine's speed was fixed at 2500 rpm. The experimental results showed that the main hydrocarbons present in WPO were within the range of diesel fuel (C13–C18, approximately 74.39%), while its specific gravity and flash point were out of the limit prescribed by the diesel fuel specification. The addition of n-butanol to WPO was found to reduce the engine's thermal efficiency and increase HC and CO emissions, especially when the engine operated at low-load conditions. In order to find the suitable ratio of n-butanol blends when the engine operated at the tested engine load, the optimization process was carried out by considering the engine's load and ratio of the n-butanol blend as input factors and the engine's performance and emissions as output factors. It was found that the multi-objective function produced by the general regression neural network (GRNN) can be modeled as the multi-objective function with high predictive performances. The coefficient of determination (R2), mean absolute percentage error (MAPE), and root mean square error (RSME) of the optimization model proposed in the study were 0.999, 2.606%, and 0.663, respectively, when brake thermal efficiency was considered, while nitrogen oxide values were 0.998, 6.915%, and 0.600, respectively. As for the results of the optimization using NSGA-II, a single optimum value may not be attained as with the other methods, but the optimization's boundary was obtained, which was established by making a trade-off between brake thermal efficiency and nitrogen oxide emissions. According to the Pareto frontier, the engine load and ratio of the n-butanol blend that caused the trade-off between maximum brake thermal efficiency and minimum nitrogen oxides are within the approximate range of 37 N.m to 104 N.m and 9% to 14%, respectively.

SA-FGDEM: A Self-adaptive E-Learning Performance Prediction Model

With the rapid development of computer computing power and the severe challenges brought by the COVID-19, e-learning, as the optimal solution for most students and other learner groups, plays an extremely important role in maintaining the normal operation of educational institutions. As the user community continues to expand, it has become increasingly important to guarantee the quality of teaching and learning. One way to ensure the quality of online education is to construct e-learning behavior data to build learning performance predictors. Still, most studies have ignored the intrinsic correlation between e-learning behaviors. Therefore, this study proposes an adaptive feature fusion-based e-learning performance prediction model (SA-FGDEM) relying on the theoretical model of learning behav-ior classification. The experimental results show that the feature space mined by fine-grained differential evolution algorithm and the adaptive feature fusion combined with differential evolution algorithm can support e-learning performance prediction more effectively and is better than the benchmark method. © 2022 IEEE.

A Review on the Learner's Performance Prediction Techniques in MOOC Courses through Data Mining

Nowadays, Massive Open Online Courses are in demand owing to their informative value, easy access and low costs. The Covid-19 pandemic era saw a lot of teaching and learning through the online resources. One of the developing fields is Educational Data Mining in which the data derived from the educational environments is collected in databases, which is further analyzed to extract some interesting patterns of information. The findings can aid in supporting the educational staff in designing a cohort that may produce better results in terms of increasing the learner's performance, identifying at-risk students, placement prediction and dropout prediction, whatever the current motive may be. In this paper, we emphasize on the techniques focusing on the performance prediction that have been applied during the years 2012 to 2022 and the attributes affecting the performance have been determined. © 2022 IEEE.

Academic performance prediction associated with synchronous online interactive learning behaviors based on the machine learning approach

With long-term impact of COVID-19 on education, online interactive live courses have been an effective method to keep learning and teaching from being interrupted, attracting more and more attention due to their synchronous and real-time interaction. However, there is no suitable method for predicting academic performance for students participating in online class. Five machine learning models are employed to predict academic performance of an engineering mechanics course, taking online learning behaviors, comprehensive performance as input and final exam scores (FESs) as output. The analysis shows the gradient boosting regression model achieves the best performance with the highest correlation coefficient (0.7558), and the lowest RMSE (9.3595). Intellectual education score (IES) is the most important factor of comprehensive performance while the number of completed assignment (NOCA), the live viewing rate (LVR) and the replay viewing rate (RVR) of online learning behaviors are the most important factors influencing FESs. Students with higher IES are more likely to achieve better academic performance, and students with lower IES but higher NOCA tend to perform better. Our study can provide effective evidences for teachers to adjust teaching strategies and provide precise assistance for students at risk of academic failure in advance.

Predicting students' academic performance using e-learning logs

The outbreak of coronavirus disease 2019 (COVID-19) drives most higher education systems in many countries to stop face-to-face learning. Accordingly, many universities, including Jordan University of Science and Technology (JUST), changed the teaching method from face-to-face education to electronic learning from a distance. This research paper investigated the impact of the e-learning experience on the students during the spring semester of 2020 at JUST. It also explored how to predict students' academic performances using e-learning data. Consequently, we collected students' datasets from two resources: the center for e-learning and open educational resources and the admission and registration unit at the university. Five courses in the spring semester of 2020 were targeted. In addition, four regression machine learning algorithms had been used in this study to generate the predictions: random forest (RF), Bayesian ridge (BR), adaptive boosting (AdaBoost), and extreme gradient boosting (XGBoost). The results showed that the ensemble model for RF and XGBoost yielded the best performance. Finally, it is worth mentioning that among all the e-learning components and events, quiz events had a significant impact on predicting the student's academic performance. Moreover, the paper shows that the activities between weeks 9 and 12 influenced students' performances during the semester.

Academic performance prediction associated with synchronous online interactive learning behaviors based on the machine learning approach

With long-term impact of COVID-19 on education, online interactive live courses have been an effective method to keep learning and teaching from being interrupted, attracting more and more attention due to their synchronous and real-time interaction. However, there is no suitable method for predicting academic performance for students participating in online class. Five machine learning models are employed to predict academic performance of an engineering mechanics course, taking online learning behaviors, comprehensive performance as input and final exam scores (FESs) as output. The analysis shows the gradient boosting regression model achieves the best performance with the highest correlation coefficient (0.7558), and the lowest RMSE (9.3595). Intellectual education score (IES) is the most important factor of comprehensive performance while the number of completed assignment (NOCA), the live viewing rate (LVR) and the replay viewing rate (RVR) of online learning behaviors are the most important factors influencing FESs. Students with higher IES are more likely to achieve better academic performance, and students with lower IES but higher NOCA tend to perform better. Our study can provide effective evidences for teachers to adjust teaching strategies and provide precise assistance for students at risk of academic failure in advance. [ FROM AUTHOR]

Identify Students at Risk Based on Behavioural Patterns in Continuous Assessment

Students' success is the ultimate goal of any institution around the world. Early detection of at-risk students can facilitate the instructor or tutor to provide timely support to those at risk of failing the course. In a traditional face-to-face classroom, students can monitor learning patterns in routine interactions. However, teachers in the online classroom have limited information, compared with the face-to-face classroom, to detect students in trouble due to the lack of instance interactions between teachers and students. Particularly, such a problem has become worse than ever since 2020, as online teaching and learning are ubiquitous in the Post-COVID19 Era. In this work, we aim to predict if the student obtains a low course grade based on their behavioral patterns in continuous assessments, which are easy-to-retrieve attributes and available in most e-learning systems. We leverage the ratio of assessment grade to the time spent on the assessment as a useful feature in the machine-learning prediction framework. Experiments on real-world datasets indicate that such a ratio can improve the accuracy of detecting at-risk students. © 2022 IEEE.

Students' performance-prediction-Model based on Physical and Physiological Constraints

The consequences of the Covid-19 pandemic changed the education system and the lifestyle of all students in Jordan. To reduce the infection rate among students, the education institutes in Jordan decided to adopt online learning as an alternative to face-to-face education. The fast shift to online education raises a potent concern regarding its efficiency. For instance, many students in Jordan cannot afford digital tools and do not have an internet connection. Furthermore, the psychological impact of enforcing online learning is not fully recognized. This study presents two regression models based on Multilayer Perceptron (MLP) neural network and Random Forest (RF) regressor to analyze and predict students' performance in Jordan before and during the lockdown and under physical and psychological constraints. In this study, the Dataset of Jordanian University Students' Psychological Health Impacted by Using E-learning Tools during COVID-19 (JUSPH) is divided into four subsets based on their chronological timeline (Before/After Covid-19), physical and psychological states. Besides, the four subsets are pre-processed using a Simple Imputer (SI), label encoder, and on-hot encoding to impute the missing value and handle the categorical data, respectively. Then, the features are selected by using the Low Variance (LV) filter. Afterward, MLP and RF regressor is used to predict the future students' performance under online education in the following semester. Results showed that the proposed MLP models achieved the best accuracy score of 99.94% on the Before Covid-19 physical Subset, while the RF model achieved the best accuracy score of 85.58% on the After Covid-19 Psychological subset. © 2022 IEEE.

AISAR: Artificial Intelligence-Based Student Assessment and Recommendation System for E-Learning in Big Data

Educational systems have advanced with the use of electronic learning (e-learning), which is a promising solution for long-distance learners. Students who engage in e-learning can access tests and exams online, making education more flexible and accessible. This work reports on the design of an e-learning system that makes recommendations to students to improve their learning. This artificial intelligence-based student assessment and recommendation (AISAR) system consists of score estimation, clustering, performance prediction, and recommendation. In addition, the importance of student authentication is recognised in situations in which students must authenticate themselves prior to using the e-learning system using their identity, password, and personal identification number. Individual scores are determined using a recurrent neural network (RNN) based on student engagement and examination scores. Then, a density-based spatial clustering algorithm (DBSCAN) using Mahalanobis distance clustering is implemented to group students based on their obtained score values. The constructed clusters are validated by estimating purity and entropy. Student performance is predicted using a threshold-based MapReduce (TMR) procedure from the score-based cluster. When predicting student performance, students are classified into two groups: average and poor, with the former being divided into below- and above-average students and the latter into poor and very poor students. This categorisation aims to provide useful recommendations for learning. A recommendation reinforcement learning algorithm, the rule-based state–action–reward–state–action (R-SARSA) algorithm, is incorporated for evaluation. Students were required to work on their subjects according to the provided recommendations. This e-learning recommendation system achieves better performance in terms of true-positives, false-positives, true-negatives, false-negatives, precision, recall, and accuracy.

Comparison of Nitrogen Dioxide Predictions During a Pandemic and Non-pandemic Scenario in the City of Madrid using a Convolutional LSTM Network

Traditionally, machine learning technologies with the methods and capabilities available, combined with a geospatial dimension, can perform predictive analyzes of air quality with greater accuracy. However, air pollution is influenced by many external factors, one of which has recently been caused by the restrictions applied to curb the relentless advance of COVID-19. These sudden changes in air quality levels can negatively influence current forecasting models. This work compares air pollution forecasts during a pandemic and non-pandemic period under the same conditions. The ConvLSTM algorithm was applied to predict the concentration of nitrogen dioxide using data from the air quality and meteorological stations in Madrid. The proposed model was applied for two scenarios: pandemic (January–June 2020) and non-pandemic (January–June 2019), each with sub-scenarios based on time granularity (1-h, 12-h, 24-h and 48-h) and combination of features. The Root Mean Square Error was taken as the estimation metric, and the results showed that the proposed method outperformed a reference model, and the feature selection technique significantly improved the overall accuracy.

Adaptive Individual Differences in Math Courses

A higher education that can be defined as sustainable ensures the acquisition of competencies that are necessary to address the current and future needs of the society in which it exists. Because math competencies are an essential component of college students’ academic and professional success, poor performance outcomes are particularly problematic in the context of an education that aims to be sustainable. This research sought to identify dispositions that are predictive of math performance in the post-pandemic world to develop an early detection system for at-risk students of an understudied population (college students of Middle Eastern descent from Saudi Arabia). It specifically targeted female and male students in STEM or non-STEM majors who were enrolled in a math course of the general education curriculum. During the second semester of a return to entirely face-to-face instruction, their self-efficacy, math learning anxiety, math evaluation anxiety, and preference for morning or evening study activities were surveyed. In the post-pandemic world of this understudied population, the math performance of STEM male and female students was hurt by concerns about learning math. The math performance of non-STEM male students benefited from self-efficacy, whereas that of non-STEM female students was unaffected by any of the dispositions surveyed in the present investigation. These findings suggest that individual difference measures can inform early interventions intended to address performance deficiencies in selected groups of students with the overreaching goal of ensuring a sustainable education for all.

Predicting User Satisfaction of Mobile Healthcare Services Using Machine Learning: Confronting the COVID-19 Pandemic

Outbreak of the COVID-19 leads to rapid development of the mobile healthcare services. Given that user satisfaction is of great significance in inducing marketing success in competition markets, this research explores and predicts user satisfaction with mobile healthcare services. Specifically, the current research aimed to design a machine learning model that predicts user satisfaction with healthcare services using big data from Google Play Store reviews and satisfaction ratings. By dealing with the sentimental features in online reviews with five classifiers, the authors find that Logistic regression with term frequency-inverse document frequency (TF-IDF) and XGBoost with Bag of words (BoW) have superior performances in predicting user satisfaction for healthcare services. Based on these results, the authors conclude that such user-generated texts as online reviews can be used to predict user satisfaction, and Logistic regression with TF-IDF and XGBoost with BoW can be prioritized for developing online review analysis platforms for healthcare service providers.