Course Evaluation Based on Deep Learning and SSA Hyperparameters Optimization

Abstract

Sentiment analysis attracts the attention of Egyptian Decisionmakers in the education sector. It offers a viable method to assess education quality services based on the students' feedback as well as that provides an understanding of their needs. As machine learning techniques offer automated strategies to process big data derived from social media and other digital channels, this research uses a dataset for tweets' sentiments to assess a few machine learning techniques. After dataset preprocessing to remove symbols, necessary stemming and lemmatization is performed for features extraction. This is followed by several machine learning techniques and a proposed Long Short-Term Memory (LSTM) classifier optimized by the Salp Swarm Algorithm (SSA) and measured the corresponding performance. Then, the validity and accuracy of commonly used classifiers, such as Support Vector Machine, Logistic Regression Classifier, and Naive Bayes classifier, were reviewed. Moreover, LSTM based on the SSA classification model was compared with Support Vector Machine (SVM), Logistic Regression (LR), and Naive Bayes (NB). Finally, as LSTM based SSA achieved the highest accuracy, it was applied to predict the sentiments of students' feedback and evaluate their association with the course outcome evaluations for education quality purposes.

Natural Language Processing of Student's Feedback to Instructors: A Systematic Review

“…4) Non-educational data: For the application of NLP methods, some studies used non-educational data from social media or already available labelled datasets. For example, El-Demerdash et al [38] used over 16K random tweets to train their sentiment analysis algorithms. Nguyen et al [26] used the UIT -ViNames dataset with over 26K Vietnamese full names along with the course data.…”

“…Based on the findings, the literature is categorised into three according to the target data language. Note that the information regarding the data could not be extracted from four of the studies ( [17], [21], [38], [42]).…”

Henry gas solubility optimization double machine learning classifier for neurosurgical patients

“…The statistical test analysis was conducted using Wilcoxon’s test based on accuracy metric. The Wilcoxon test is a non-parametric test [ 56 ], therefore it has less assumptions than parametric tests such as t-test. As a result, the Wilcoxon test is performed when the t-test for dependent samples fails to meet its boundary criteria.…”

An Evaluation Framework for Online Courses Based on Sentiment Analysis Using Machine Learning

“…For example, Wang et al [7] proposed the lite bidirectional encoder representation from transformers (BERT) bi-directional long short-term memory (ALBERT-BiLSTM) sentiment analysis model for multiple open online course (MOOC) review comments. Alaa et al [8] used the salp swam algorithm, a long short-term memory classifier, to predict the emotions of students based on their course feedback. Qi et al [9] used latent Dirichlet allocation (LDA) to develop a course evaluation system that describes various aspects in online course review comments.…”