International Journal of Modern Education and Computer Science (IJMECS)

IJMECS Vol. 16, No. 6, Dec. 2024

Cover page and Table of Contents: PDF (size: 673KB)

Table Of Contents

REGULAR PAPERS

Exploring Holistic Well-Being: EDA and a Comparative ML Approach to Work-Life Balance

By Bala Dhandayuthapani V.

DOI: https://doi.org/10.5815/ijmecs.2024.06.01, Pub. Date: 8 Dec. 2024

This research focuses on lifestyle and work-life balance, examining how different dynamics can influence holistic wellbeing, which focuses on the current literature to develop a framework that extends beyond standard well-being concepts. The work aims to enrich the discourse on holistic well-being by synthesizing existing knowledge and exploring new knowledge, providing valuable views for individuals and age groups involved in professionals striving to develop a more balanced and meaningful life. The study utilizes AI-driven methods and machine learning algorithms to analyse work-life indices to understand patterns and correlations that contribute or hinder work-life harmony. The findings highlight the importance of lifestyle choices, social connections, and personal fulfilment in achieving holistic wellness. The research provides evidence-based insights and practical recommendations to develop a healthy lifestyle and work-life balance. The study also examines the ethical implications of AI and highlights the need for a comprehensive approach to work-life balance. The study utilizes supervised learning algorithms and a comparative analysis of the accuracy scores of various algorithms, revealing significant differences in classification and more accuracy for the work-life balance score. The study aims to uncover insights beyond the typical contrasts and illuminate the interrelationship between numerous factors affecting an individual's well-being.

[...] Read more.
Analyzing Sentiments on Twitter Using Deep Learning Techniques

By Aditya Bhushan Devanshi Dwivedi Ashutosh Kumar Singh Snehlata

DOI: https://doi.org/10.5815/ijmecs.2024.06.02, Pub. Date: 8 Dec. 2024

In today’s digital age dominated by social media, understanding public sentiment through Twitter analysis has become imperative. With a staggering 100 million active users on platforms like Twitter and an influx of 572,000 new accounts daily, the vast reservoir of user-generated content underscores the necessity for advanced sentiment analysis tools. This study delves into the realm of sentiment analysis techniques on Twitter, with a particular emphasis on employing Machine Learning (ML) methods. The proposed framework harnesses the power of Natural Language Processing (NLP) and Deep Learning architectures, specifically advocating for a synergistic blend of Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM) networks. Additionally, it explores the efficacy of traditional ML algorithms such as Support Vector Machines (SVM), Random Forest, and Multi-Layer Perceptron (MLP) in this context. The study’s findings illuminate diverse performance metrics across the employed models. While SVM exhibits moderate accuracy, it grapples with challenges in recall and F1-score for sentiment class 1. Conversely, the CNN-LSTM model emerges as a standout performer, boasting impressive accuracy rates of 97% and 98% respectively. Notably, this model excels in sentiment classification across all classes, underscoring its efficacy in discerning nuanced sentiment nuances within tweets. Furthermore, the study underscores the critical importance of judiciously selecting ML algorithms tailored to the intricacies of Twitter sentiment analysis. By leveraging advanced NLP techniques and deep learning architectures, researchers and practitioners can glean deeper insights into the dynamic landscape of public sentiment on social media platforms like Twitter. Such insights hold significant implications for diverse domains, including marketing, brand management, and public opinion analysis.

[...] Read more.
Emotion Classification Utilizing Transformer Models with ECG Signal Data

By Ch. Raga Madhuri Kundu Bhagya Sri Kasaraneni Gagana Tiprineni Sathvika Lakshmi

DOI: https://doi.org/10.5815/ijmecs.2024.06.03, Pub. Date: 8 Dec. 2024

In recent years, there has been growing interest in leveraging physiological signals, such as Electrocardiogram (ECG) data, for emotion classification tasks. This study explores the efficacy of utilizing Transformer models, a state-of-the-art architecture in natural language processing, for emotion classification using ECG signal data. The proposed methodology involves preprocessing the ECG signals, extracting relevant features, and model architecture consists of DistilBERT model, Pooling Layer to obtain a fixed-size representation of the ECG signal, Dropout Layer to prevent overfitting, Fully Connected Layer for classification. Experiments are conducted on publicly available dataset, demonstrating the effectiveness of the proposed approach compared to traditional machine learning methods. The results suggest that DistilBERT Transformer model can effectively capture complex temporal dependencies within ECG signals, thereby achieving notable performance of 76% accuracy in emotion classification tasks. This research contributes to the growing body of literature exploring the intersection of physiological signals and deep learning techniques for affective computing applications.

[...] Read more.
A Prediction Method of Forecasting University Student Achievements Using an Iterative Neural Network Shrinking Algorithm

By Amir Abdul Majid

DOI: https://doi.org/10.5815/ijmecs.2024.06.04, Pub. Date: 8 Dec. 2024

The aim of this work is to predict high education students’ progress and achievement by forecasting their final grades in any taught courses as early as possible during a study semester or term, using an innovative neural network shrinking technique. The trained neural network NN is divided into a head and several tails in a cascaded sequential manner. Course data from previous offerings are used in a network which is trained using a feedforward BP scheme to extract input and output weight coefficients and biases. One of the input features is reduced by a fraction of its original value and used with the same input data in a tailed NN, which is initiated with the extracted coefficients and biases from the previous run. The training is continued in a cascaded manner until eliminating one input assignment. The whole process is continued for other assignments to be eliminated. This algorithm can be constituted as a dynamic process workbench for an alternative method of forecasting the achieved grades of high educational students in an easy and cost-free manner. The earlier it is to forecast final grade the easier it is to alleviate outcomes of course grading. The procedure is applied on two different courses offered by the same teacher. The input data of different batches of students attending a particular course are used. It is found that a tentative accuracy of predicting final grades from the start is possible.

[...] Read more.
Impacts of AR in Learning Tribal Bodo Language

By Dipali Basumatary Debojeet Das Ranjan Maity

DOI: https://doi.org/10.5815/ijmecs.2024.06.05, Pub. Date: 8 Dec. 2024

The objective of this study is to determine the effect of augmented reality compared to traditional approaches for language learning in primary education.  We reported BodoAR, a marker-based AR application suitable for teaching the Bodo language, a tribal language of India in this work. The proposed application was developed using UNITY. To find the efficacy of the proposed BodoAR, five research questions were formed. An empirical study was conducted in five primary schools on two distinct user groups: native and non-native speakers of the Bodo language. All the participants were again randomly divided into two other groups- the experimental and the control group. A mixed-method approach was employed to collect the data, utilizing quantitative and qualitative methodologies. Analysis of empirical data shows native Bodo speakers' learning - in terms of academic achievements was significantly improved using the proposed application. For non-native students, experimental groups performed better than the controlled group. Both the native and non-native experimental groups experienced low anxiety, positive attitudes, and high levels of satisfaction while using the BodoAR application. Additionally, the academic achievements and attitudes of the native students in the experimental group were positively and significantly correlated. In contrast, the achievement of non-native students exhibits a positive and significant relationship with usability. The findings show increased satisfaction and academic performance among students who used the BodoAR application, affirming its effectiveness in enhancing Bodo language learning for primary school children. Thus, AR can be a useful tool for incorporating into children’s language learning for children.

[...] Read more.
The Correlation of Podcast Storytelling Duration with Discussion Timing in Enhancing EFL Learners' Listening Skills

By Mohamed A. Elkot Rabea Ali Mohammed AbdAlgane Eltaieb Youssif Walid Aboraya

DOI: https://doi.org/10.5815/ijmecs.2024.06.06, Pub. Date: 8 Dec. 2024

Recently, podcasts have gained attention as an experimental tool for enhancing English language skills. However, existing studies often overlook crucial variables related to learning dynamics. The current research aims to fill this gap by investigating the impact of podcast duration, the timing of educational content discussion, and their interplay on developing English language listening skills. The study explores the nexus between tale duration and discussion timing across four distinct groups via the quasi-experimental design. These groups undergo varied levels of correlation between story duration and discussion timing, allowing the research to scrutinise their effects on English as a Foreign Language (EFL) learners' listening skills. The study's sample comprises sixty university students majoring in English language and translation. Rigorous participant selection criteria, including language proficiency, educational background, and prior experience with podcast-based language learning, ensure a diverse yet homogeneous group. The randomly assigned participants form four experimental groups, each exposed to different combinations of story duration and discussion timing. Additionally, this research provides valuable insights into the intricate relationship between these variables and their impact on EFL learners' listening skills. Findings from cognitive assessments and observation cards reveal that, surprisingly, these factors do not significantly influence students' English listening competence ratings. However, irrespective of the schedule, pre-lecture discussions emerge as a consistently effective strategy for enhancing students' English listening skills. Moreover, in light of these findings, the study presented a set of recommendations that emphasized the importance of discussion and interaction among students before listening to digital content, which in turn reflects positively on the performance of academic students in developing English as a foreign language listening skills.

[...] Read more.
Predicting Education Level of the Farmers‟ Children of a Developing Country during COVID 19 Using Machine Learning

By Md. Mehedi Rahman Rana Md. Nasim Adnan Md. Moradul Siddique Md. Tahadur Rahman Ferdib-Al-Islam

DOI: https://doi.org/10.5815/ijmecs.2024.06.07, Pub. Date: 8 Dec. 2024

Education is one of the necessities of an individual’s life, as it enhances the self-morality and nobility that leads one towards the challenging pathways of the competitive world. In the agricultural based country, education is scarce among the children of the farmers as they suffer from poverty. After affecting with COVID-19, study dropout rate of farmers’ children is increased. We collected raw data from rural areas of different countries, and pre-processed this data before applying the machine learning algorithm to improve the performance. We used advanced machine learning models to predict whether farmer’s children will run or drop out of their education. Based on the outcomes it was viewed that, machine learning strategies substantiate to be suitable in this area. This research proposes preventive steps for dropping out of the farmers' children. It also shows that, the Random Forest being the highest reliable model for foreseeing dropout rate and education level. 

[...] Read more.