Article Per Year (5 Year)
p-Index From 2021 - 2026
0.61
P-Index
This Author published in this journals
All Journal Jurnal INFOTEL Research in Education, Technology, and Multiculture
Shoffan Saifullah
AGH University of Krakow, Poland
Computer Science & IT Electrical & Electronics Engineering

Published : 3 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 3 Documents
Search

 1 
The Evaluation of Effects of Oversampling and Word Embedding on Sentiment Analysis Nur Heri Cahyana; Yuli Fauziah; Wisnalmawati Wisnalmawati; Agus Sasmito Aribowo; Shoffan Saifullah
JURNAL INFOTEL Vol 17 No 1 (2025): February 2025
Publisher : LPPM INSTITUT TEKNOLOGI TELKOM PURWOKERTO

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.20895/infotel.v17i1.1077

Abstract

Generally, opinion datasets for sentiment analysis are in an unbalanced condition. Unbalanced data tends to have a bias in favor of classification in the majority class. Data balancing by adding synthetic data to the minority class requires an oversampling strategy. This research aims to overcome this imbalance by combining oversampling and word embedding (Word2Vec or FastText). We convert the opinion dataset into a sentence vector, and then an oversampling method is applied here. We use 5 (five) datasets from comments on YouTube videos with several differences in terms, number of records, and imbalance conditions. We observed increased sentiment analysis accuracy with combining Word2Vec or FastText with 3 (three) oversampling methods: SMOTE, Borderline SMOTE, or ADASYN. Random Forest is used as machine learning in the classification model, and Confusion Matrix is used for validation. Model performance measurement uses accuracy and F-measure. After testing with five datasets, the performance of the Word2Vec method is almost equal to FastText. Meanwhile, the best oversampling method is Borderline SMOTE. Combining Word2Vec or FastText with Borderline SMOTE could be the best choice because of its accuracy score and F-measure reaching 91.0% - 91.3%. It is hoped that the sentiment analysis model using Word2Vec or FastText with Borderline SMOTE can become a high-performance alternative model.
Semi-Supervised Sentiment Classification Using Self-Learning and Enhanced Co-Training Agus Sasmito Aribowo; Siti Khomsah; Shoffan Saifullah
JURNAL INFOTEL Vol 17 No 3 (2025): August
Publisher : LPPM INSTITUT TEKNOLOGI TELKOM PURWOKERTO

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.20895/infotel.v17i3.1344

Abstract

Sentiment classification is usually done manually by humans. Manual senti- ment labeling is ineffective. Therefore, automated labeling using machine learning is es- sential. Building a computerized labeling model presents challenges when labeled data is scarce, which can decrease model accuracy. This study proposes a semi-supervised learn- ing (SSL) framework for sentiment analysis with limited labeled data. The framework integrates self-learning and enhanced co-training. The co-training model combines three machine learning methods: Support Vector Machine (SVM), Random Forest (RF), and Lo- gistic Regression (LR). We use TF-IDF and FastText for feature extraction. The co-training model will generate pseudo-labels. Then, the pseudo-labels from models (SVM, RF, LR) are checked to choose the highest confidence — this is called self-learning. This framework is applied to English and Indonesian language datasets. We ran each dataset five times. The performance difference between the baseline model (without pseudo-labels) and SSL (with pseudo-labels) is not significant; the Wilcoxon Signed-Rank Test confirms it, obtaining a p- value < 0.05. Results show that SSL produces pseudo-labels on unlabeled data with quality close to the original labels on unlabeled data. Although the significance test performs well on four datasets, it has not yet surpassed the performance of the supervised classification (baseline). Labeling using SSL proves more efficient than manual labeling, as evidenced by the processing time of around 10-20 minutes to label thousands to tens of thousands of samples. In conclusion, self-learning in SSL with co-training can effectively label unla- beled data in multilingual and limited datasets, but it has not yet converged across various datasets.
A Serious Game with Augmented Reality to Improve Cognitive Learning of Computer Networks Dwi Ratnawati; Vivianti Vivianti; Yuwono Indro Hatmojo; Shoffan Saifullah
Research in Education, Technology, and Multiculture Vol 5, No 1 (2026): Research in Education, Technology, and Multiculture
Publisher : Institute of Multidisciplinary Research and Community Service

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.61436/rietm/v5i1.pp15-28

Abstract

Understanding computer networks requires a strong cognitive foundation, particularly in mastering abstract concepts such as network topology, device configuration, and data transmission processes. However, conventional learning methods are often limited in providing concrete visualization and interactive experiences. This study aims to develop and evaluate a serious game integrated with augmented reality (AR) technology to improve students’ cognitive learning outcomes in computer networks. This research employed a development approach by integrating serious game design with AR-based visualization. The development process involved asset creation, learning objective formulation, scenario design, and the implementation of problem-based cases on computer network concepts. The resulting product allows students to interact with virtual network devices in a real-world environment via mobile devices, enabling flexible learning anytime, anywhere. The evaluation phase included expert validation, user testing, and effectiveness testing. The results indicated that the developed media is feasible and effective for learning. Media expert validation reached 73.50%, while material expert validation achieved 87.50%, both categorized as highly feasible. User responses showed a feasibility level of 85.26%, indicating positive acceptance. Furthermore, the effectiveness test demonstrated a moderate improvement in students’ cognitive learning outcomes, with a gain percentage of 64.41%. The quantitative scores were obtained by converting Likert-scale ratings into percentages using a standardized formula to ensure transparency and consistency in data analysis. In conclusion, the integration of serious games and augmented reality provides an interactive and contextual learning experience that supports cognitive understanding of computer networks. This approach is considered both feasible and sufficiently effective for implementation in vocational education settings, particularly in technology-based learning environments. Keywords: Computer network learning, Serious game, Augmented reality, Learning media development, Effectiveness evaluation.
Co-Authors Agus Sasmito Aribowo Agus Sasmito Aribowo Dwi Ratnawati, Dwi Nur Heri Cahyana Siti Khomsah Vivianti Wisnalmawati Wisnalmawati Yuli Fauziah Yuwono Indro Hatmojo, Yuwono