Educational data mining: a systematic review of research and emerging trends

Du, Xu; Yang, Juan; Hung, Jui-Long; Shelton, Brett E.

doi:10.1108/idd-09-2019-0070

Cited by 36 publications

(21 citation statements)

References 52 publications

(68 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Referensi [12] menyajikan tinjauan sistematis mengenai penelitian terkait EDM dan tren yang berkembang. Dalam penelitian tersebut dinyatakan bahwa teknik ML bahkan DL telah digunakan secara luas pada konteks EDM dan prediksi kinerja murid merupakan salah satu topik penelitian utama [12]. Penelitian lain memaparkan tinjauan mengenai analisis dan prediksi kinerja murid melalui penggunaan ML [13].…”

Section: Pendahuluanunclassified

Tinjauan Pustaka Sistematis: Implementasi Metode Deep Learning pada Prediksi Kinerja Murid

Diponegoro

Kusumawardani

Hidayah

2021

JNTETI

View full text Add to dashboard Cite

Pemanfaatan machine learning yang merupakan salah satu implementasi dalam bidang artificial intelligence telah merambah ke berbagai bidang, salah satunya adalah bidang pendidikan. Dengan menggunakan kombinasi teknik machine learning, statistik, dan basis data, dapat dilakukan educational data mining untuk mengetahui pola yang ada dalam suatu dataset tertentu. Salah satu penggunaan educational data mining adalah untuk melakukan prediksi kinerja murid. Hasil dari prediksi kinerja murid dapat digunakan sebagai salah satu instrumen untuk melakukan monitoring dan evaluasi terhadap proses pembelajaran sehingga dapat membantu menentukan langkah-langkah lanjutan dalam rangka meningkatkan proses pembelajaran. Makalah ini bertujuan untuk mengetahui state of the art implementasi deep learning yang merupakan bagian dari machine learning pada konteks educational data mining, khususnya mengenai prediksi kinerja murid. Pada makalah ini disajikan systematic literature review untuk mengetahui variasi teknik atau algoritme deep learning yang digunakan beserta kinerja yang dicapai. Dari dua puluh publikasi ilmiah yang ditelusuri, rata-rata kinerja yang dicapai dalam melakukan prediksi adalah 89,85%. Mayoritas teknik yang digunakan adalah Deep Neural Network (DNN), Recurrent Neural Network (RNN), dan Long Short-Term Memory (LSTM) dengan fitur data demografis, perilaku, dan akademis.

show abstract

Section: Pendahuluanunclassified

Tinjauan Pustaka Sistematis: Implementasi Metode Deep Learning pada Prediksi Kinerja Murid

Diponegoro

Kusumawardani

Hidayah

2021

JNTETI

View full text Add to dashboard Cite

show abstract

“…The learner’s learning performance is an important indicator to evaluate his/her learning outcomes, so the difference in learning performance is often used to prove the impacts of learning preferences (Du et al , 2020). Studies in this category were conducted by experimentation.…”

Section: The Interaction Between Learning Preference and Other Factors In A Learning Contextmentioning

confidence: 99%

Learning preference: development in smart learning environments

Zhang

Hung

et al. 2021

IDD

Self Cite

View full text Add to dashboard Cite

Purpose The purpose of this study is to conduct a systematic review to understand state-of-art research related to learning preferences from the aspects of impacts, influential factors and evaluation methods. Design/methodology/approach This paper uses the systematic synthesis method to provide state-of-the-art knowledge on learning preference research by summarizing published studies in major databases and attempting to aggregate and reconcile the scientific results from the individual studies. The findings summarize aggregated research efforts and improve the quality of future research. Findings After analyzing existing literature, this study proposed three possible research directions in the future. First, researchers might focus on how to use the real-time tracking mechanism to further understand other impacts of learning preferences within the learning environments. Second, existing studies mainly focused on the influence of singular factors on learning preferences. The joint effects of multiple factors should be an important topic for future research. Finally, integrated algorithms might become the most popular evaluation method of learning preference in the era of smart learning environments. Research limitations/implications This review used the search results generated by Google Scholar and Web of Science databases. There might be published papers available in other databases that have not been taken into account. Originality/value The research summarizes the state-of-art research related to learning preferences. This paper is one of the first to discuss the development of learning preference research in smart learning environments.

show abstract

“…As previously mentioned, Du et al [13] adopted this strategy, in which the arbiter tree was created in a bottom-up fashion [25]. For the most part, the dataset is arbitrarily truncated into many sub-partitions of size 'n'.…”

Section: Arbiter Treesmentioning

confidence: 99%

“…In another research study, the authors proposed a prediction system based on the Adaboost algorithm to reduce risk failure by providing timely advice to high-risk students [10]. Additional study was done on educational data to increase prediction accuracy using different ensemble methodologies [11][12][13]. The authors also recommended several prediction paradigms based on their findings.…”

Section: Introductionmentioning

confidence: 99%

“…When various classifiers are hybridized together, a large number of outputs are formed by voting among the participants. According to Du et al [13], an uneven learning process is required in order to generate statistically meaningful findings [17]. The basic classifier is also trained repeatedly on additional samples of the original dataset utilizing random sampling with a replacement technique in this method, which is a significant improvement over previous methods.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

An Ensemble and Filtering-Based System for Predicting Educational Data Mining

Hananto¹

2021

J. Appl. Data Sci.

View full text Add to dashboard Cite

When developing a prediction paradigm, an ensemble technique such as boosting is used. It is built on a heuristic framework. Generally speaking, engineering ensemble learning is more accurate than individual classifiers when it comes to making predictions. Consequently, numerous ensemble strategies have been presented in this work, particularly to provide a more complete understanding of the essential methods in general. Researchers have experimented with boosting methods to forecast student performance as part of a variety of ensemble techniques. The researchers employed improvement approaches to construct an accurate predictive educational model, which was based on a key phenomena seen in categorization and prediction operations. In light of the uniqueness and originality of the suggested strategy in educational data mining, the researchers used augmentation strategies in order to construct an accurate predictive pedagogical model. Tenfold cross-validation was performed to evaluate the effectiveness of the basic classifiers, which included the random tree, the j48, the knn, and the Naive Bayes. The random tree was found to be the most effective classifier. Several additional screening techniques, including oversampling (SMOTE) and undersampling (Spread subsampling), were utilized to analyze any statistically significant variations in results between the meta and base classifiers that had been identified between the meta and base classifiers. The use of ensemble and screening strategies, as compared to the use of standard classifiers, has demonstrated considerable gains in predicting student performance, as has the use of either strategy alone. Furthermore, after the completion of a performance research on each approach, two new prediction models have been established on the basis of the improved results gained thus far.

show abstract

Educational data mining: a systematic review of research and emerging trends

Cited by 36 publications

References 52 publications

Tinjauan Pustaka Sistematis: Implementasi Metode Deep Learning pada Prediksi Kinerja Murid

Tinjauan Pustaka Sistematis: Implementasi Metode Deep Learning pada Prediksi Kinerja Murid

Learning preference: development in smart learning environments

An Ensemble and Filtering-Based System for Predicting Educational Data Mining

Contact Info

Product

Resources

About