Fig 1 - uploaded by Anup Majumder
Content may be subject to copyright.
Source publication
Nowadays, In Bangladesh, the dropout rate at post-graduation level or incompletion of the post-graduation degree is considered as a serious problem in the education sector. This work can be used to support for identifying the specific individuals as well as the institutional factors which may next lead to the enrollment or drop out at the post-grad...
Similar publications
The detection of coronavirus (COVID-19) is now a critical task for the medical practitioner. The coronavirus spread so quickly between people and approaches 100,000 people worldwide. In this consequence, it is very much essential to identify the infected people so that prevention of spread can be taken. In this paper, the deep learning based method...
Citations
... Online education has revolutionized the world of teaching and learning, providing new opportunities to access education from anywhere in the world [14], [15]. In recent years, ML has started to play a very important role in online education, enhancing the learning experience and personalizing education for each student [16], [17]. ML is a discipline within the field of artificial intelligence (AI) that provides the ability for computers to learn and improve their performance on a specific task without being explicitly programmed. ...
... One of the most common of these refers to the risk of dropout, either from a full program or from a single course. Such predictions attempt to identify which students might not complete the academic path they initiated and to support these students until successful completion (Biswas et al., 2019;Borrella et al., 2019;Cohen, 2017;Yukselturk et al., 2018). Another approach aims at predicting student achievement; that is, identifying an at-risk behaviour at a lower threshold. ...
In this qualitative study (N=6), we explored insights of first-year students' instructors and advisors into an early identification system aimed at detecting non-thriving students in the context of an all-campus first-year orientation course for undergraduates. Following the development of that prediction model in a bottom-up manner, using a plethora of available data, we focus on how its end-users could help us understand the underlying mechanisms that drive the identification of non-thriving students. As findings suggest, participants were appreciative overall of the prediction and its timing and came up with various behaviours that could explain non-thriving, mostly motivation and engagement. They suggested additional data that could predict non-thriving, including background information, academic engagement, and learning habits. Notes for Practice • Change terminology regarding early identification systems from survive to thrive, to reflect students' potential realization • Combine quantitative and qualitative approaches for model construction to achieve reliable, authentic learning analytics • Set institutional policies regarding a wide range of data types, to support effective data-driven decision-making
... The main advantages of RF are that each tree protects each other from their own individual errors. The working principle of the random forest model is shown in Fig. 4 [15]. In this paper, we have used the ensemble based technique to combine the predictions from multiple machine learning algorithms together to make more accurate predictions than any individual model. ...
Online Education has become a buzzword since the
COVID-19 hit the World. Most of the educational institutions
went online to continue educational activities while developing
countries like Bangladesh took a significant period of time
to ensure online education at every education level. Students
of several levels also faced many difficulties when they got
introduced to online education. It is important for the decision makers
of educational institutions to be informed about the
effectiveness of online education so that they can take further
steps to make it more beneficial for the students. Our main
motivation is to contribute to this matter by analyzing the
relevant factors associated with online education. In this work,
we have collected students’ information of all three different
levels(School, College, and University) by conducting both online
and physical surveys. The surveys form consists of an individual’s
socio-demographic factors. To get an idea about the effectiveness
of online education we have applied several machine learning
algorithms named Decision Tree (DT), Random Forest (RF),
Naive Bayes (NB), Support Vector Machine (SVM), K-Nearest
Neighbors (KNN), and also Artificial Neural Network(ANN) on
our dataset to predict the adaptability level of the students to
online education. Among used algorithms, the Random Forest
classifier achieved the best accuracy of 89.63% and outperformed
other algorithms.
... Finally, they found SVM outperforms others by achieving 85.76% accuracies. Biswas et al. [13] applied seven classifiers namely Multilayer Perceptron, Naïve Bayes, Logistic, Random Forest, Locally Weighted Learning, Random Tree, and Part. They have calculated the seven metrics to evaluate their work and measure the performance of the classifier. ...
Nowadays, the job market has become more competitive due to the advancement of higher educations. In Bangladesh, an individual selects three types of careers namely Government Job, Private Job, and Business after the completion of their graduation. Selecting a specific career is affected by many factors. These factors can be personal as well as academic. In this research, we have done career predictions and analyze several factors behind this. To accomplish the work, we have analyzed several academic and personal factors. Here, we have applied several data mining techniques for experimentation and several performance evaluation metrics to evaluate our work. Finally, we found that Part classifier outperforms the other techniques.
... Biswas et al. [17] used diverse machine learning classifiers to predict the enrollment and dropout status in the postgraduation level. For this work, they collected the dataset from a renowned public university of Bangladesh. ...
Every year thousands of students get admitted into different universities in Bangladesh. Among them, a large number of students complete their graduation with low scoring results which affect their careers. By predicting their grades before the final examination, they can take essential measures to ameliorate their grades. This article has proposed different machine learning approaches for predicting the grade of a student in a course, in the context of the private universities of Bangladesh. Using different features that affect the result of a student, seven different classifiers have been trained, namely: Support Vector Machine (SVM), K-Nearest Neighbor (KNN), Logistic Regression, Decision Tree, AdaBoost, Multilayer Perceptron (MLP), and Extra Tree Classifier for classifying the students' final grades into four quality classes: Excellent, Good, Poor, and Fail. Afterwards, the outputs of the base classifiers have been aggregated using the weighted voting approach to attain better results. And here this study has achieved an accuracy of 81.73%, where the weighted voting classifier outperforms the base classifiers.
... They stated that the logistic regression model works better than other models to predict total undergraduate enrollment. Biswas et al. [9], applied seven machine learning classifiers to predict the dropout and enrollment in the post-graduation degree. First, they computed the confusion matrix for each of the classifiers. ...
Bangladesh is a densely populated country where a large portion of citizens is living under poverty. In Bangladesh, a significant portion of higher education is accomplished at private universities. In this twenty-first century, these students of higher education are highly mobile and different from earlier generations. Thus, retaining existing students has become a great challenge for many private universities in Bangladesh. Early prediction of the total number of registered students in a semester can help in this regard. This can have a direct impact on a private university in terms of budget, marketing strategy, and sustainability. In this paper, we have predicted the number of registered students in a semester in the context of a private university by following several machine learning approaches. We have applied seven prominent classifiers, namely SVM, Naive Bayes, Logistic, JRip, J48, Multilayer Perceptron, and Random Forest on a data set of more than a thousand students of a private university in Bangladesh, where each record contains five attributes. First, all data are preprocessed. Then preprocessed data are separated into the training and testing set. Then, all these classifiers are trained and tested. Since a suitable classifier is required to solve the problem, the performances of all seven classifiers need to be thoroughly assessed. So, we have computed six performance metrics, i.e. accuracy, sensitivity, specificity, precision, false positive rate (FPR) and false negative rate (FNR) for each of the seven classifiers and compare them. We have found that SVM outperforms all other classifiers achieving 85.76% accuracy, whereas Random Forest achieved the lowest accuracy which is 79.65%.
Student dropout is non-attendance from school or college for an extended period for no apparent cause. Tending to this issue necessitates a careful comprehension of the basic issues as well as an appropriate intervention strategy. Statistical approaches have acquired much importance in recent years in resolving the issue of student dropout. This is due to the fact that statistical techniques can efficiently and effectively identify children at risk and plan interventions at the right time. Thirty-six studies in total were reviewed to compile, arrange, and combine current information about statistical techniques applied to predict student dropout from various academic databases between 2000 and 2023. Our findings revealed that the Random Forest in 23 studies and the Decision Tree in 16 studies were among the most widely adopted statistical techniques. Accuracy and Area Under the Curve were the frequently used evaluation metrics that are available in existing studies. However, it is notable that the majority of these techniques have been developed and tested within the context of developed nations, raising questions about their applicability in different global settings. Moreover, our meta-analysis estimated a pooled proportion of overall dropouts of 0.2061 (95% confidence interval: 0.1845–0.2278), revealing significant heterogeneity among the selected studies. As a result, this systematic review and meta-analysis provide a brief overview of statistical techniques focusing on strategies for predicting student dropout. In addition, this review highlights unsolved problems like data imbalance, interpretability, and geographic disparities that might lead to new research in the future.
Nowadays in Bangladesh, one of the most common scenarios for our students is moving to other countries for higher studies. In order to succeed, students need to pick the right direction before applying for a higher education visa. This article aims to implement which student’s visa will get approved or rejected for their higher studies abroad by using machine learning. Throughout this analysis, we predict the visa for higher studies based on student data. Then we process the data (such as cleaning, transformation, integration, standardization, and feature selection). We used multiple classification methods later on, i.e., C4.5 (j48), k-NN, naive Bayes, random forest, SVM, neural network, for classifying these models. Depending on the outcome of the study, it was observed that the accuracy, confusion matrix, and other factors of a random forest classifier are more compatible than others. GRE score and undergraduate CGPA, are two of the most significant attributes that rely on deciding the success of visa acceptance for higher studies also discovered.
