Student achievement prediction using deep neural network from multi-source campus data

Xiao-yong, LI; Zhang, Yong; Cheng, Huimin; Li, Mengran; Yin, Baocai

doi:10.1007/s40747-022-00731-8

Cited by 20 publications

(14 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…One of the most suitable approaches for analyzing neuroscience mechanisms is using FNNs. [14,15] employ NNs for analysis to evaluate changes and receive feedback on the learning process. It can be concluded that problem-solving skills should be developed using a computerized strategy that incorporates both biological and mathematical NNs, along with a combination of neuroscience concepts such as motivation, emotion, metacognition, working memory, processing speed, execution, and memory, among others.…”

Section: Neural Network and Neuroscience Mechanistic Strategies To Pr...mentioning

confidence: 99%

See 1 more Smart Citation

An Intelligent Mathematics Problem-Solving Tutoring System Framework

Mohamad Ariffin Abu Bakar,

Ahmad Termimi Ab Ghani,

Mohd Lazim Abdullah

2024

Int. J. Onl. Eng.

View full text Add to dashboard Cite

This study proposed a novel framework for redesigning problem-solving activities in an intelligent tutoring system (ITS) called the intelligent neural-mechanistic mathematics problem- solving tutoring system (IN-MP-STS). This concept paper presents a new approach to ITS by incorporating elements of neuroscience mechanisms as a learning strategy that focuses on optimizing the brain’s ability through neural mechanisms. It also introduces fuzzy neural networks (FNNs) as a tool for modulating assessment and analyzing outcomes. This framework offers an alternative perspective on delivery methods and learning approaches in the ITS module. By effectively integrating neuroscience mechanistic elements such as motivation, activation, regulation, execution, memorization, and interactivities, deep learning can be achieved, leading to improved student competence. This framework also proposes an adaptive assessment component based on FNNs, which will enhance the measurement and feedback modules in the system. It is necessary to modify the way that ITS and soft computing methods, such as the study of neural networks (NNs), are combined to make learning measurement and assessment more transparent. This innovation has not been fully disclosed, so researchers are encouraged to further test the concepts presented to assess their alignment with the existing system and ethical considerations. This framework enhances the conceptual research findings of FNNs and incorporates neuroscience-based strategies into architecture and autonomous problem-solving skills within an ITS model. It also offers references for the development of problem-solving learning. IN-MP-STS has the potential to significantly enhance students’ competencies and abilities, thereby fostering the development of more comprehensive, holistic, and sustainable ITS. This approach also has the potential to enrich the existing literature on the sustainability of neural networks.

show abstract

Section: Neural Network and Neuroscience Mechanistic Strategies To Pr...mentioning

confidence: 99%

“…The complexity of the data source that describes learning necessitates the use of an appropriate structure or architecture for the NN algorithm. In this context, fuzzy NNs (FNNs) were chosen because of their adaptability to a variety of orientations and data sources, as evidenced by several researchers [12][13][14][15][16][17].…”

Section: Introductionmentioning

confidence: 99%

An Intelligent Mathematics Problem-Solving Tutoring System Framework

Mohamad Ariffin Abu Bakar,

Ahmad Termimi Ab Ghani,

Mohd Lazim Abdullah

2024

Int. J. Onl. Eng.

View full text Add to dashboard Cite

show abstract

“…The experiment showed that decision trees had the best performance in predicting English scores, and Bayesian algorithms had the best performance in predicting mathematics scores, with an accuracy rate of 87.1% and 83.9%, respectively, and found that there was a certain relationship between mathematics and English scores. Xiaoyong Li et al [5] proposed an end-to-end deep learning model that could automatically extract features from heterogeneous student behavior data to predict learning outcomes. They collected and experimented with daily behavior data from 8228 students, and the experiment showed that the accuracy of this deep learning model in grade prediction was better than traditional machine learning algorithms.…”

Section: Related Workmentioning

confidence: 99%

Multi-Output Based Hybrid Integrated Models for Student Performance Prediction

Xue

Niu

2023

Applied Sciences

View full text Add to dashboard Cite

In higher education, student learning relies increasingly on autonomy. With the rise in blended learning, both online and offline, students need to further improve their online learning effectiveness. Therefore, predicting students’ performance and identifying students who are struggling in real time to intervene is an important way to improve learning outcomes. However, currently, machine learning in grade prediction applications typically only employs a single-output prediction method and has lagging issues. To advance the prediction of time and enhance the predictive attributes, as well as address the aforementioned issues, this study proposes a multi-output hybrid ensemble model that utilizes data from the Superstar Learning Communication Platform (SLCP) to predict grades. Experimental results show that using the first six weeks of SLCP data and the Xgboost model to predict mid-term and final grades meant that accuracy reached 78.37%, which was 3–8% higher than the comparison models. Using the Gdbt model to predict homework and experiment grades, the average mean squared error was 16.76, which is better than the comparison models. This study uses a multi-output hybrid ensemble model to predict how grades can help improve student learning quality and teacher teaching effectiveness.

show abstract

“…The data set for this research was obtained from Beijing University over 145 days in the spring semester. Using extract, transform, and load (ETL) techniques, 9000 students' campus behaviors (Li et al 2022) were gathered from several databases. Data, time, place, and consumption quantity are the additional four characteristics of consumer behavior.…”

Section: Data Elucidationmentioning

confidence: 99%

Analysis of educational data enabled by deep learning to increase student success

Kumar,

Chaudhary,

Choudhary

2023

Multidiscip. Sci. J.

View full text Add to dashboard Cite

A key component of improving educational quality is identifying pupils who are at a high risk of doing poorly academically as early as feasible. To accomplish this, most studies now in existence have used conventional Deep learning (DL) algorithms to forecast students' academic progress based on their behavior data, from which behavior elements are manually identified owing to the professional expertise and knowledge. Nevertheless, it has become increasingly difficult to recognize finely constructed handcrafted traits as a result of a rise in the types and quantities of behavioral data. The Enriched Plant Growth Optimized Artificial Neural Network (EPGO-ANN) technique enabled data analysis of educational data which is a viable tactic that may be used to improve student accomplishment in educational settings as we suggested in this research. The optimization predicts the academic success by autonomously extracting characteristics from student behavior data from several heterogeneous sources. This model's novelty employs recording the in-built time-series data for each kind of activity and uses EPGO-ANN to extract correlation features between various behaviors. The results of the experiments show that the suggested deep model approach outperforms several DL methods. The results of experiments show that the EPGO-ANN technique is superior to other DL methods.

show abstract

Student achievement prediction using deep neural network from multi-source campus data

Cited by 20 publications

References 34 publications

An Intelligent Mathematics Problem-Solving Tutoring System Framework

An Intelligent Mathematics Problem-Solving Tutoring System Framework

Multi-Output Based Hybrid Integrated Models for Student Performance Prediction

Analysis of educational data enabled by deep learning to increase student success

Contact Info

Product

Resources

About