Garuda - Garba Rujukan Digital

Dede Sunandar

Pamulang University

Author-ID : 9843858

Computer Science & IT

Published : 3 Documents Claim Missing Document

Claim Missing Document

Articles

Thesis Title Similarity Detection System Using Levenshtein Distance and Cosine Similarity Dede Sunandar; Adam Muiz
bit-Tech Vol. 8 No. 1 (2025): bit-Tech
Publisher : Komunitas Dosen Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32877/bt.v8i1.2864

The manual verification process of thesis titles in higher education institutions is often time-consuming and prone to oversight, making it difficult to ensure the uniqueness of each student’s work. This poses serious academic risks, as undetected similarities in thesis titles can lead to unintended plagiarism, compromise academic integrity, and undermine the credibility of educational institutions. In a broader sense, repeated or overlapping research topics also reflect a lack of innovation and weaken the scientific contribution of academic programs. To address this issue, an automated detection system is needed to efficiently identify similarities between thesis titles. This study aims to develop a web-based thesis title similarity detection system that integrates Levenshtein Distance and Cosine Similarity algorithms. The system was developed using the Waterfall model, involving stages of requirements analysis, design, implementation, and evaluation. Functional features such as login, title data management, old spelling normalization, and real-time similarity detection were implemented. The results show that the combination of both algorithms effectively detects similarities in character and semantic aspects. The inclusion of an old spelling normalization feature significantly improves detection accuracy by aligning historical and modern word forms prior to analysis. In conclusion, the developed system not only supports a faster and more objective title verification process but also contributes to the prevention of academic plagiarism and promotes integrity in higher education environments.

Web-Based Citrus Fruit Disease Detection Application Using MobileNet V2 for Agricultural Quality Assurance Nasrul Hidayah; Adam Muiz; Dede Sunandar
bit-Tech Vol. 8 No. 2 (2025): bit-Tech
Publisher : Komunitas Dosen Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32877/bt.v8i2.3338

Precision disease detection in citrus commodities has become increasingly essential within the framework of Agriculture 4.0, particularly for small-scale vendors who still rely on manual visual inspection that is often inconsistent and error-prone. This study develops and evaluates a web-based citrus fruit disease detection system using the MobileNet V2 Convolutional Neural Network architecture. The methodological novelty of this work lies in the integration of an optimized MobileNet V2 model enhanced through targeted data augmentation and lightweight fine-tuning into an end-to-end web ecosystem that supports two inference modes: static image upload and real-time webcam-based detection, tailored to the operational needs of small vendors. The system classifies citrus fruit images into four categories: Black-Spot, Citrus Canker, Greening (Huanglongbing), and Fresh, using more than 1,000 augmented images standardized to 224×224 pixels with an 80:20 train–test split. Experimental results show that the model achieves an accuracy of 96.21%, with consistently high precision and recall across disease classes, while the Fresh class exhibits relatively higher misclassification due to visual similarity with early-stage symptoms. The Flask-based web application demonstrates stable performance under black-box testing and delivers rapid, high-confidence predictions. These findings affirm the effectiveness of lightweight CNN approaches in improving fruit quality inspection accuracy, reducing sorting errors, and supporting more efficient workflows for local vendors.

Identification of Dengue Hemorrhagic Fever (DHF) Using the Naïve Bayes Classifier Method Erdy Sutriyatna; Dede Sunandar; Adam Muiz
bit-Tech Vol. 8 No. 2 (2025): bit-Tech
Publisher : Komunitas Dosen Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32877/bt.v8i2.3378

Dengue Hemorrhagic Fever (DHF) remains a major public health burden in tropical and subtropical countries, with Indonesia consistently reporting the highest incidence in Southeast Asia since 1968. Early diagnosis traditionally depends on clinical evaluation and laboratory confirmation, processes that may require several days and often delay intervention during the critical plasma leakage phase. Addressing this gap, the present study introduces an intelligent early identification system for DHF based on the Naïve Bayes Classifier, a probabilistic data-mining method recognized for its computational efficiency and strong performance in handling categorical medical attributes. The model was trained using 100 anonymized patient records and DHF screening forms collected from Puskesmas Pasir Buah, Curug, Bitung, spanning 2020–2023, incorporating twelve clinically relevant predictors consisting of symptom-based indicators and basic hematological parameters. Following preprocessing and 10-fold cross-validation, the system achieved an average accuracy of 94.67%, precision of 95.2%, recall of 93.8%, and an F1-score of 94.5%, demonstrating its reliability for preliminary DHF assessment. The resulting web-based prototype allows health workers to input patient symptoms and receive immediate probabilistic classifications (Positive/Negative) accompanied by recommendations for confirmatory laboratory testing. By providing rapid and interpretable diagnostic support, this system has the potential to reduce diagnostic delays at the primary healthcare level and enhance decision-making in resource-limited environments.

Title Search

Found 3 Documents Search

Abstract

Abstract

Abstract

Title

Found 3 Documents
Search