Student Performance Prediction Research Articles

Cloud-based machine learning algorithms for anomalies detection

Gradient boosting machines harnesses the inherent capabilities of decision trees and meticulously corrects their errors in a sequential fashion, culminating in remarkably precise predictions. Word2Vec, a prominent word embedding technique, occupies a pivotal role in natural language processing (NLP) tasks. Its proficiency lies in capturing intricate semantic relationships among words, thereby facilitating applications such as sentiment analysis, document classification, and machine translation to discern subtle nuances present in textual data. Bayesian networks introduce probabilistic modeling capabilities, predominantly in contexts marked by uncertainty. Their versatile applications encompass risk assessment, fault diagnosis, and recommendation systems. Gated recurrent units (GRU), a variant of recurrent neural networks, emerges as a formidable asset in modeling sequential data. Both training and testing are crucial to the success of an intrusion detection system (IDS). During the training phase, several models are created, each of which can recognize typical from anomalous patterns within a given dataset. To acquire passwords and credit card details, "phishing" usually entails impersonating a trusted company. Predictions of student performance on academic tasks are improved by hyper parameter optimization of the gradient boosting regression tree using the grid search approach.

Machine learning based education data mining through student session streams

Recently, significant growth in using online-based learning stream (i.e., elearning systems) have been seen due to pandemic such as COVID-19. Forecasting student performance has become a major task as an institution is focusing on improving the quality of education and students' performance. Data mining (DM) employing machine learning (ML) techniques have been employed in the e-learning platform for analyzing student session streams and predicting academic performance with good effects. A recent, study shows ML-based methodologies exhibit when data is imbalanced. In addressing ensemble learning by combining multiple ML algorithms for choosing the best model according to data. However, the existing ensemblebased model does not incorporate feature importance into the student performance prediction model. Thus, exhibits poor performance, especially for multi-label classification. In addressing this, this paper presents an improved ensemble learning mechanism by modifying the XGBoost algorithm, namely modified XGBoost (MXGB). The MXGB incorporates an effective cross-validation scheme that learns correlation among features more efficiently. The experiment outcome shows the proposed MXGBabased student performance prediction model achieves much better prediction accuracy contrary to the state-of-art ensemble-based student performance prediction model.

Student Performance Prediction

Abstract: Academic integration is now defined by how students are able to access, and effectively interact with, and are supported by engineering faculty and staff; Academic integration, which is a very important construct that contributes to a student’s persistence and ultimate graduation. But, several factors, either numeric or non-numeric can have varying degrees of influence on academic performance of students. Rankings of aforementioned factors, hopefully, will assist the management in order to initiate corrective action with a view to students' academic performance improvement. In an academic environment, teachers are frequently in close contact with students and therefore their belief about the degree of influence of factors on students’ academic performance could be of vital significance. The case study presented in this paper uses fuzzy relational calculus and the Dempster Shafer theory of evidence to examine and suggest a feasible and realistic ranking approach of only chosen factors!

Risk‐based student performance prediction model for engineering courses

AbstractHigh academic failure and dropout rates in engineering courses are significant worldwide concerns attributed to various factors, with academic performance being a critical variable. This article provides a methodology to estimate the performance risk of students in engineering schools. Risk analysis is a strategy to evaluate academic success, which provides a set of methods to analyze, understand, and predict student outcomes before enrolling in specific majors or challenging college courses. This article develops a methodology to estimate fragility curves for students entering an engineering course. The fragility function concept, borrowed from the earthquake engineering field, estimates the likelihood of success in a course, given relevant student metadata, such as the grade point average, thus comprehensively addressing student performance variability. A student academic success prediction model enables instructional designers to make informed decisions. For example, fragility curves can help achieve two goals: (i) assessing the population at risk for a course to take actions to improve student success rates and (ii) assessing a course's relative difficulty based on its fragility function parameters. We demonstrate this methodology through a case study comparing the relative difficulty of two engineering courses, Statics and Solid Mechanics, at a university in Colombia. Given that Statics serves as a prerequisite for Solid Mechanics, deficiencies in the former can significantly impact student performance in the latter. The case study results reveal that Solid Mechanics poses a higher risk of academic failure than Statics, underscoring the importance of a strong foundation in prerequisite courses.

A Novel Deep Learning Model for Student Performance Prediction Using Engagement Data

Technology-enhanced learning supported by virtual learning environments (VLEs) facilitates tutors and students. VLE platforms contain a wealth of information that can be used to mine insight regarding students’ learning behaviour and relationships between behaviour and academic performance, as well as to model data-driven decision-making. This study introduces a system that we termed ASIST: a novel Attention-aware convolutional Stacked BiLSTM network for student representation learning to predict their performance. ASIST exploits student academic registry, VLE click stream, and midterm continuous assessment information for their behaviour representation learning. ASIST jointly learns the student representation using five behaviour vectors. It processes the four sequential behaviour vectors using a separate stacked bidirectional long short term memory (LSTM) network. A deep convolutional neural network models the diurnal weekly interaction behaviour. It also employs the attention mechanism to assign weight to features based on their importance. Next, five encoded feature vectors are concatenated with the assessment information, and, finally, a softmax layer predicts the high-performer (H), moderate-performer (M), and at-risk (F) categories of students. We evaluate ASIST over three datasets from an Irish university, considering five evaluation metrics. ASIST achieves an area under the curve (AUC) score of 0.86 to 0.90 over the three datasets. It outperforms three baseline deep learning models and four traditional classification models. We also found that the attention mechanism has a slight impact on ASIST’s performance. The ablation analysis reveals that weekly event count has the greatest impact on ASIST, whereas diurnal weekly interaction has the least impact. The early prediction using the first seven weeks of data achieves an AUC of 0.83 up to 0.89 over the three datasets. In yearly analysis, ASIST performs best over the 2018/19 dataset and worst over the 2020/21 dataset.

Prediction of Engineering Students Learning Outcomes using Machine Learning

Data science and machine learning, over the years have proven very well-organized and significant in many sectors including education. Machine learning is an aspect of artificial intelligence in which a computing system can able to learn from data and make conclusions. The recent development in education sector provides assessment tools to predict the student performance by exploring education data using machine learning and data mining techniques. Student performance assessment is an important measurement metrics in education which affects the university accreditation. Student performance improvement plan must be implemented in those universities, by counselling the low performer students. It helps both students and teachers to overcome the problems experienced by the student during studies and teaching techniques of teachers. In this review paper, different student performance prediction literature related to find out low performer student. The survey results indicated that different machine learning techniques are used to overcome the problems related to predicting student at risk and assessment of student performance. Machine learning techniques plays an important role in progress and prediction of student performance, thus improving student performance prediction system

Prediction of student exam performance using data mining classification algorithms

AbstractStudent outcomes are of great importance in higher education institutions. Accreditation bodies focus on them as an indicator to measure the performance and effectiveness of the institution. Forecasting students’ academic performance is crucial for every educational establishment seeking to enhance performance and perseverance of its students and reduce the failure rate in the future. The main goal of this study is to predict the performance of undergraduate first-level students in the Computer Department during the years 2016 to 2021 to enhance their performance in future by discovering the best algorithm use to analyze the educational data to identify the students’ academic performance. The secondary data was collected by reviewing the Student Affairs Department at the Faculty of Specific Education at Damietta University, in addition to the Statistics Department at the university. The dataset contained 830 instances after excluding 139 instances of missing values, irrelevant rows, and outliers. The dataset was divided into train (577 instances (70%)), test (253 instances (30%)) and involved six features such year, midterm, practical exam, writing exam, final total degree, and grade. This paper use five machine learning (ML) algorithms which was selected according to the literature review and high accuracy in predicting educational data mining: For the purpose of comparison, a number of different machine learning algorithms, such as Random Forest, Decision Tree, Naive Bayes, Neural Network, and K-Nearest Neighbours, were utilized and evaluated with evaluation metrics such as confusion matrix, accuracy, precision, recall, and F-measure. The Random Forest and Decision Tree classifiers emerged as the top-performing algorithms, accurately categorizing 250 instances when predicting students' performance in the statistics course. This was determined based on the findings of the study. Out of a total of 253 instances that were included in the testing set, they only made three incorrect classifications.

Early Prediction of Students Performance in Higher Education

Evaluating students' learning performance is a fundamental aspect of evaluating any educational institution. When addressing challenges related to the learning process, student performance is critical, and it is one of the key factors used to quantify learning outcomes. The topic of research known as educational data mining (EDM) has grown out of the potential to leverage data knowledge to enhance educational systems. EDM is the development of methods to analyze data collected from educational environments, enabling a more complete and precise understanding of students and the enhancement of their educational results. Evaluating the students' learning results is a crucial part of evaluating any educational institution. One of the key variables used to quantify learning outcomes is student performance, which is significant when addressing problems with the learning process. The field of research known as educational data mining, or EDM, was born out of the potential to leverage data knowledge to enhance educational systems. EDM is the process of developing methods for evaluating information obtained from educational environments. This makes it possible to learn more precise and in-depth information about students and enhances their academic achievement. Academic achievement tests (AAT), general aptitude tests (GAT), admission scores, first-level courses, and other early-stage factors are used in the paper's dimensionality reduction mechanism by T-SNE algorithm for the clustering technique. This allows the study to investigate the relationship between these aspects and GPAs. Regarding the categorization method, the study showcases tests conducted on various machine learning models that forecast student achievement in the initial phases by utilizing diverse attributes such as course grades and entrance exam results. To gauge the models' quality, we employ various evaluation measures. Based on the findings, it appears that early student failure rates can be reduced by educational institutions.

Prediction of Student Performance Using Random Forest Combined With Naïve Bayes

Abstract Random forest is a powerful ensemble learning technique celebrated for its heightened predictive performance and robustness in handling complex datasets; nevertheless, it is criticized for its computational expense, particularly with a large number of trees in the ensemble. Moreover, the model’s interpretability diminishes as the ensemble’s complexity increases, presenting challenges in understanding the decision-making process. Although various pruning techniques have been proposed by researchers to tackle these issues, achieving a consensus on the optimal strategy across diverse datasets remains elusive. In response to these challenges, this paper introduces an innovative machine learning algorithm that integrates random forest with Naïve Bayes to predict student performance. The proposed method employs the Naïve Bayes formula to evaluate random forest branches, classifying data by prioritizing branches based on importance and assigning each example to a single branch for classification. The algorithm is utilized on two sets of student data and is evaluated against seven alternative machine-learning algorithms. The results confirm its strong performance, characterized by a minimal number of branches.

Student Performance Prediction Using Machine Learning Algorithms

Education is crucial for a productive life and providing necessary resources. With the advent of technology like artificial intelligence, higher education institutions are incorporating technology into traditional teaching methods. Predicting academic success has gained interest in education as a strong academic record improves a university’s ranking and increases student employment opportunities. Modern learning institutions face challenges in analyzing performance, providing high-quality education, formulating strategies for evaluating students’ performance, and identifying future needs. E-learning is a rapidly growing and advanced form of education, where students enroll in online courses. Platforms like Intelligent Tutoring Systems (ITS), learning management systems (LMS), and massive open online courses (MOOC) use educational data mining (EDM) to develop automatic grading systems, recommenders, and adaptative systems. However, e-learning is still considered a challenging learning environment due to the lack of direct interaction between students and course instructors. Machine learning (ML) is used in developing adaptive intelligent systems that can perform complex tasks beyond human abilities. Some areas of applications of ML algorithms include cluster analysis, pattern recognition, image processing, natural language processing, and medical diagnostics. In this research work, K-means, a clustering data mining technique using Davies’ Bouldin method, obtains clusters to find important features affecting students’ performance. The study found that the SVM algorithm had the best prediction results after parameter adjustment, with a 96% accuracy rate. In this paper, the researchers have examined the functions of the Support Vector Machine, Decision Tree, naive Bayes, and KNN classifiers. The outcomes of parameter adjustment greatly increased the accuracy of the four prediction models. Naïve Bayes model’s prediction accuracy is the lowest when compared to other prediction methods, as it assumes a strong independent relationship between features.

Early Performance Prediction of Students in Higher Education

Abstract: An essential component of any evaluation of an educational institution is assessing the learning outcomes of its pupils. Student performance is important when addressing issues with the learning process and is one of the main variables used to quantify learning outcomes. The area of study referred to as the idea of using data knowledge to improve educational systems gave rise to educational data mining, or EDM. The creation of techniques for analyzing data gathered from educational settings is known as EDM. This allows for a more thorough and accurate knowledge of pupils as well as an improvement in their academic performance. Assessing the learning outcomes of students is an essential component of assessing any educational establishment. When addressing issues with the learning process, student performance is an important metric utilized to quantify learning outcomes

Educational data mining for student performance prediction: feature selection and model evaluation

Student performance is a multidimensional concept that includes academic accomplishment and cognitive development. It measures the performance of educational institutions, teaching methods and individual efforts. Engagement, motivation and support networks have a substantial impact on performance results. In this research, we intend to establish an intelligent data mining-based model for predicting student performance through a novel feature selection approach. We propose an innovative Adaptive Sea Horse Optimization (ASHO) to select the crucial features to predict the education performance level. We gathered a xAPI-Edu-Data dataset which includes students’ numerous academic details, to train our suggested model. The Z-score normalization algorithm is employed to pre-process the obtained raw data, it improves the quality of the data. We utilized Independent Component Analysis (ICA) to extract relevant characteristics from the processed data. We utilize the ASH algorithm for feature selection, it dynamically adjusts search parameters, efficiently exploring the feature space to locate the ideal subset of features for improving the predictive performance. The selected features are classified by the implementation of the eXtreme Gradient Boosting (XGBoost) algorithm for predicting student performance. Our recommended approach is implemented in Python software. The finding evaluation phase examines the suggested model's prediction effectiveness with various parameters such as precision, accuracy, recall and f1 score. We performed a comparison analysis with other traditional methods to determine the effectiveness of the proposed approach. The experimental results demonstrate that the proposed prediction approach performed better than other existing approaches.

THE FUTURE OF ENGINEERING EDUCATION: A DATA ANALYTICS APPROACH

The landscape of engineering education is undergoing a significant transformation, driven by the integration of data analytics and machine learning (ML) technologies. This paper explores the potential of these technologies to revolutionize engineering education by personalizing learning experiences, predicting student performance, and enhancing curriculum development. Through a comprehensive review of current practices and case studies, we examine the application of data-driven approaches in identifying individual learning patterns, tailoring educational content, and implementing adaptive learning technologies. We also investigate the role of predictive analytics in forecasting academic success and enabling proactive interventions for at-risk students. Furthermore, the paper discusses the challenges and ethical considerations associated with the adoption of these technologies, including data privacy concerns and the digital divide. Our analysis highlights the importance of collaboration among educators, policymakers, and technologists to navigate these challenges and fully realize the benefits of a data-analytics approach in engineering education. The paper concludes with a vision for the future, emphasizing the need for continuous innovation and adaptation in curricula to prepare engineering graduates for the evolving demands of the workforce and society. This investigation sheds light on the transformative potential of data analytics and ML in engineering education and provides a roadmap for its successful integration into teaching and learning processes.
 Keywords: Data Analytics, Machine Learning, Personalized Learning, Predictive Analytics, Curriculum Development, Adaptive Learning Technologies, Student Performance Prediction, Technological Integration.

Framelet-based dual hypergraph neural networks for student performance prediction

Framelet-based dual hypergraph neural networks for student performance prediction

Language model-guided student performance prediction with multimodal auxiliary information

Student Performance Prediction (SPP) has received a lot of attention due to its educational implications, such as personalized instruction. Among numerous attempts in SPP, recommendation-based approaches (e.g. matrix factorization; MF) are especially attractive because of the specialized nature of personalization. However, they commonly struggle with the cold-start problem and incorporating side information. While existing works cope with the problem by utilizing additional information, such as students’ personalities, none of them cover the multimodal auxiliary information that can unleash the full potential of the recommendation-based SPP. In this work, we leverage multimodal auxiliary information at SPP for the first time by adopting large language models (GPT-J and Llama 2) as information blenders to produce extra guidance signals for the MF method. Specifically, our language model-guided matrix factorization (LMgMF), consumes the verbalized multimodal information, produces semantically rich embeddings for educational interaction, and uses them as auxiliary signals for MF. While doing so, we harness a API-level black-box language model without requiring parameter accessibility to adapt them. We evaluate LMgMF on two real-world datasets: (1) a large-scale Korean dataset that contains 1.6M instances from 10K pre-high school students; (2) ASSISTments2009 covering 346K interactions between students and math questions. Throughout extensive validation, we demonstrate that LMgMF consistently outperforms baseline methods in various SPP scenarios, including the cold-start.

PLACEMENT PREDICTION SYSTEM USING MACHINE LEARNING

Engineering students are not sure what they want to study after graduation. Students are confused by the many options offered by universities such as postgraduate admissions, and factors such as salaries and different jobs worsen the situation. There is no reliable platform that allows students to predict outcomes from the beginning of engineering and take action to bridge the gap and create a better future. Students studying in engineering faculties need to know where they stand compared to others and what kind of placement they will get. Training and workshops are available when students enter their final year, but these are not useful for students to plan their future studies. Student placement is one of the most important goals of the school. Schools work hard to accommodate students. The aim is to predict the current year's student placement by analyzing the data students have collected from previous years. Prediction of student performance is an important part of the study because today all student development is directly related to the student's success in tests and activities. Therefore, there are many situations where it is necessary to predict student performance, for example, identifying students who are not performing well and taking steps to improve them. There is no platform for girls to review their current work and highlight their strengths. . Currently, there are platforms that are not trained on real and complete data and cannot learn from error prediction, which reduces accuracy in the long run. Our goal is to create one. To ensure the results are acceptable, the model will be trained on real data and a large number of positive and negative results will be obtained. The model proposes an algorithm to estimate the model. Information is collected by the school where the prediction will be made, using the necessary information before the process or by examining the historical data of the previous year's students, and the placement of the current students is predicted and helps to improve the placement percentage of the students. institute. Keywords: task planning, task prediction, data decision making, machine learning algorithms, prediction prediction, student prediction, good study, student learning results, pre-process data, learning results.

Bibliometric insights into data mining in education research: A decade in review

This bibliometric study on data mining in education synonymous with big educational data utilizes VOSviewer and Harzing’s Publish and Perish to analyze the metadata of 1,439 journal articles found in Scopus from 2010 to 2022. As bibliometric analyses in this field are lacking, this study aims to provide a comprehensive outlook on the current developments and impact of research in this field. This study employs descriptive and trends analysis, co-authorship analysis, co-citation analysis, co-occurrences of keywords, terms map analysis, and analysis of the impact and performance of publications. It also partially replicates a similar study conducted by Wang et al. (2022), who used the Web of Science (WoS) database. The study is reported in an article entitled ‘Big data and data mining in education: A bibliometrics study from 2010 to 2022’. Results show that data mining in education is a growing research field. There is also a significant difference between the publications in Scopus and WoS. The study found several research areas and topics, such as student academic performance prediction, e-learning, machine learning, and innovative data mining techniques, to be the core basis for collaborating and continuing current research in this field. These results highlight the importance of continuing research on data mining in education, guiding future research in tackling educational challenges.

Exploring the role of machine learning in forecasting student performance in education: An in-depth review of literature

The literature on the role of machine learning in forecasting student performance in education lacks comprehensive data quality, equitable and interpretable models, consideration of contextual and causal factors, and the integration of human expertise. This review will explore machine learning types, algorithms, predictive performance, and impact on student performance in education. A systematic literature review based on articles published in the past 2019–2023 period Methods: The IEEE Xplore database was searched by using keywords such as "Educational Data Mining," "Student Performance Prediction," "Evaluations of Students," "Performance Analysis of Students," and "Learning Curve Prediction." were employed and 50 papers were selected. Results: The analysis of the results highlighted prominent patterns. Half of the studies favored supervised learning methods, with decision trees leading (19 instances), followed by Long Short-Term Memory and Random Forest (16 each), and K-Nearest Neighbor and Naive Bayes (12 and 11 times). Support Vector Machine and Logistic Regression were noted 10 and 9 times, respectively. Noteworthy were ANN, CNN, and Xgboost. Positive impacts were evident in 36 cases; only one showed negative effects, while 13 indicated intricate relationships. This study helped in understanding the prevalent machine learning methods used for predicting student performance, provides a benchmark for assessing the effectiveness of new or alternative techniques. Conclusion: This review highlights the varied machine learning uses in predicting student performance in education, emphasizing supervised methods, diverse algorithms, and complex intervention impacts.

Enhancing Student Performance Prediction on Learnersourced Questions with SGNN-LLM Synergy

Learnersourcing offers great potential for scalable education through student content creation. However, predicting student performance on learnersourced questions, which is essential for personalizing the learning experience, is challenging due to the inherent noise in student-generated data. Moreover, while conventional graph-based methods can capture the complex network of student and question interactions, they often fall short under cold start conditions where limited student engagement with questions yields sparse data. To address both challenges, we introduce an innovative strategy that synergizes the potential of integrating Signed Graph Neural Networks (SGNNs) and Large Language Model (LLM) embeddings. Our methodology employs a signed bipartite graph to comprehensively model student answers, complemented by a contrastive learning framework that enhances noise resilience. Furthermore, LLM's contribution lies in generating foundational question embeddings, proving especially advantageous in addressing cold start scenarios characterized by limited graph data. Validation across five real-world datasets sourced from the PeerWise platform underscores our approach's effectiveness. Our method outperforms baselines, showcasing enhanced predictive accuracy and robustness.

Predictors of student preparedness for advanced pharmacy practice experiences

BackgroundA capstone course often serves as the final checkpoint of student readiness before the commencement of experiential training. The purpose of this study was to determine if the assessment components from the capstone course can serve as predictors of student performance during their Advanced Pharmacy Practice Experiences (APPEs).MethodsStudent data was analyzed to observe the correlation between performance in the Pharmacy Curriculum Outcomes Assessment (PCOA), student performance in the capstone course, and the overall grade point average (GPA) earned during APPEs. Spearman rank correlation analysis, multiple linear regression, and Mann-Whitney U test were used for statistical analysis.ResultsA statistically significant positive correlation was observed between the overall APPE GPA and students’ capstone course grade, top drug competency exam score, pharmacy calculation competency exam score, and PCOA exam score. A significant regression equation was obtained during the analysis: (F(5, 97) = 5.62, p < 0.001), with an R2 = 0.225 (adjusted R2 = 0.185). In the linear regression model, capstone GPA emerged as a significant predictor (β = 0.155; p = 0.019) of APPE GPA amongst the tested variables. Additionally, students scoring < 73% on the top drug competency exam in the capstone course or less than the reference group in the PCOA exam were found to have significantly lower GPA during their APPEs compared to other students.ConclusionPerformance on the top drug competency exam and the PCOA exam can serve as potential predictors of success during APPEs.