1 10 18 https://repository.horizon.ac.id/files/original/4b6f0a142cdbaaf97a1da5652d63c0b2.pdf cb7632d0bb07a683fdbd6eb98b00fcc7 Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Title A name given to the resource Vol 8 No 6 (2024) Text A resource consisting primarily of words for reading. Examples include books, letters, dissertations, poems, newspapers, articles, archives of mailing lists. Note that facsimiles or images of texts are still of the genre Text. Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Title A name given to the resource Lightweight Models for Real-Time Steganalysis: A Comparisonof MobileNet, ShuffleNet, and EfficientNet Subject The topic of the resource deep learning; lightweight; steganography;steganalysis; security Description An account of the resource In the digital age, the security of communication technologies is paramount, with cybercrime projected to reach $10.5 trillion annually by 2025. While encryption is vital, decrypted data remains vulnerable, prompting the exploration of steganography as an additional security layer. Steganography conceals data within digital media, but its misuse for cyberattacks—such as embedding malware—has highlighted the need for steganalysis, the detection of hidden data. Despite extensive research, few studies have explored lightweight deep-learningmodels for real-time steganalysis in resource-constrained environments like mobile devices. This research evaluates MobileNet, ShuffleNet, and EfficientNet for such tasks, using the BOSSbase-1.01 dataset. Models were assessed based on accuracy, computational efficiency, and resource usage. MobileNet achieved the highest computational speed but with only 63.8% accuracy, falling short of practical application. ShuffleNet and EfficientNetperformed at random-guessing levels with50% accuracy, reflecting the challenges of steganalysis on mobile platforms. Future work aims to improve accuracy by integrating advanced preprocessing techniques, attention mechanisms, and hybrid architectures, as well as leveraging ensemble methods for improved detection. Data augmentation, transfer learning, and hyperparameter tuning will also be explored to optimize model performance. This study contributes by identifying these challenges and offering insights for future research, focusing on optimizing models and preprocessing techniques to enhance detection accuracy in resource-constrained environments Creator An entity primarily responsible for making the resource Achmad Bauravindah1*, Dhomas Hatta Fudholi Source A related resource from which the described resource is derived https://jurnal.iaii.or.id/index.php/RESTI/article/view/6091/990 Publisher An entity responsible for making the resource available Master Program inInformatics, Faculty of Industrial Technology, Islamic University of Indonesia, Yogyakarta, Indonesia Date A point or period of time associated with an event in the lifecycle of the resource 26-12-2024 Contributor An entity responsible for making contributions to the resource FAJAR BAGUS W Format The file format, physical medium, or dimensions of the resource PDF Language A language of the resource ENGLISH Type The nature or genre of the resource TEXT https://repository.horizon.ac.id/files/original/58b664c27acf7dc5441fc4cd345be85d.pdf cbafccef8fd67b6221112431b94c5e69 Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Title A name given to the resource Vol 8 No 6 (2024) Text A resource consisting primarily of words for reading. Examples include books, letters, dissertations, poems, newspapers, articles, archives of mailing lists. Note that facsimiles or images of texts are still of the genre Text. Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Title A name given to the resource Ingredients Identification Through Label Scanning Using PaddleOCR and ChatGPT for Information Retrieval Subject The topic of the resource food identification, optical character recognition, PaddleOCR, information retrieval, ChatGPT Description An account of the resource Human health depends on choosing food ingredients that align with dietary needs and avoid allergens. However, consumers often encounter unfamiliar ingredients that require additional information. Traditionally, they search online by typing in the ingredient's name which can be time-consuming and may not yield relevant results. Therefore, a system to identify and display ingredient information is necessary. This studyproposesa new system that identifies ingredients by scanning the composition label on packaging using PaddleOCR and retrieving information through ChatGPT on a smartphone. The process begins with capturing an image of the composition label.ThenPaddleOCR is employed to extract text from the scanned label, enabling identification of the listed ingredients. Subsequently, ChatGPT retrieves detailed information about the desired ingredients and displays it, allowing users to easily understand the ingredients. The system's effectiveness in text recognition is assessed using the character error rate (CER). The results show robustperformance by achieving an average CER of 0.14, with flat packaging reaching an impressive CER of 0.05. Additionally, the system's usability was assessed through pilot testing which received significant positive user feedbackachieving a 4.37 satisfaction level on the Likert scale, particularly regarding the clarity and relevance of the ingredient information provided Creator An entity primarily responsible for making the resource Ahmad Wahyu Rosyadi1*, Siti Ma’shumah2, Muhammad Qomaruz Zaman3, Moh.Rizki Fajar Source A related resource from which the described resource is derived https://jurnal.iaii.or.id/index.php/RESTI/article/view/6119/992 Publisher An entity responsible for making the resource available Department of Informatics Engineering, Faculty of Engineering, Universitas Qomaruddin, Gresik, Indonesia Date A point or period of time associated with an event in the lifecycle of the resource 27-12-2024 Contributor An entity responsible for making contributions to the resource FAJAR BAGUS W Format The file format, physical medium, or dimensions of the resource PDF Language A language of the resource ENGLISH Type The nature or genre of the resource TEXT https://repository.horizon.ac.id/files/original/6752696f75d1c51eba1e5c22a7c0672a.pdf ab0f4aae286755ec2ea4b38ee27f39bf Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Title A name given to the resource Vol 8 No 6 (2024) Text A resource consisting primarily of words for reading. Examples include books, letters, dissertations, poems, newspapers, articles, archives of mailing lists. Note that facsimiles or images of texts are still of the genre Text. Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Title A name given to the resource Optimizing Indonesian-Sundanese Bilingual Translation with Adam-Based Neural Machine Translation Subject The topic of the resource Adam; BLEU; Indonesian; LSTM; Sundanese Description An account of the resource This research seeks to construct an automatic translation between Indonesian and Sundanese languages based on the Neural Machine Translation (NMT) method. The model used in this study is the Long Short-Term Memory (LSTM) type, which carries out an encoder-decoder structure model learned with Bible data. The text translation here was conducted in different epochs to optimize the process, followed bythe Adam optimization algorithm. Testing the Adam optimizer with different epoch settings yields a BLEU score for Indonesian to Sundanese translations of 0.991785, higher than the performance of the None optimizer. Experimental results demonstrate that Indonesian to Sundanese translation using Adam optimization with 1000 epochs consistently performed better in BLEU -Bilingual Evaluation Understudy -scoring thanSundanese to Indonesian translation. Limitations of the research were also put forth, particularly technical issues related to the collection of data and the Sundanese language’s complex grammatical features, that the model can onlypartially express, honorifics, and the problem of polysemy. Also, it mustbe mentioned that no special hyperparameter selection was performed, as parameters were chosen randomly. In future studies, transformer-based models can be investigated since these architectures will better dealwith complex language via their self-attention mechanism. Creator An entity primarily responsible for making the resource Anita Qotrun Nada1*, Aji Prasetya Wibawa2, Dhea Fanny Putri Syarifa3, Erliana Fajarwati4,Fadia Irsania Putri Source A related resource from which the described resource is derived https://jurnal.iaii.or.id/index.php/RESTI/article/view/6116/989 Publisher An entity responsible for making the resource available Faculty of Engineering, Universitas Negeri Malang, Malang, Indonesia Date A point or period of time associated with an event in the lifecycle of the resource 24-12-2024 Contributor An entity responsible for making contributions to the resource FAJAR BAGUS W Format The file format, physical medium, or dimensions of the resource PDF Language A language of the resource ENGLISH Type The nature or genre of the resource TEXT https://repository.horizon.ac.id/files/original/76cf35b046f282713d6086a455b162bb.pdf db90e7d18dc001b317e9afcd2e6d208d Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Title A name given to the resource Vol 8 No 6 (2024) Text A resource consisting primarily of words for reading. Examples include books, letters, dissertations, poems, newspapers, articles, archives of mailing lists. Note that facsimiles or images of texts are still of the genre Text. Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Title A name given to the resource Data Clustering for Sentiment Classification with Naïve Bayes and Support Vector Machine Subject The topic of the resource sentiment analysis;hotel; clustering; naïve bayes; support vector machine Description An account of the resource Visitor reviews play a crucial role in determining the success of a business, particularly those offering hospitality and services, such as hotels. The growth of internet technology has made it easier for guests to share their experiences, which can influence potential customers. Google Maps is one of the platforms used for giving and searching reviewsThis research usesdata crawled from Google Maps Review using the playwright library. However, the large volume of reviews can make analysis and topic-based categorization—such as service quality, hotel location, and operational hours—challenging. To address this, DBSCAN is used to cluster reviews based on these topics. Clustering helps improve sentiment classification, making it more targeted and allowing a comparison of two machine learning algorithms: Naïve Bayes and Support Vector Machine (SVM). Naïve Bayes achieved higher accuracy(0.87) in the operational hours cluster, while SVM scored 0.78. However, SVM showed improved accuracy in the location (0.89) and service (0.88) clusters, with Naïve Bayes maintaining a stable 0.86 accuracy in both. Both models demonstrated an average training time of less than one second, excluding preprocessing. Creator An entity primarily responsible for making the resource Bayu Yanuargi1*, Ema Utami2, Kusrini3,Arli Aditya Parikesit4 Source A related resource from which the described resource is derived https://jurnal.iaii.or.id/index.php/RESTI/article/view/6139/998 Publisher An entity responsible for making the resource available Department of Magister of Informatics Engineering, UniversitasAMIKOM Yogyakarta, Yogyakarta, Indonesia Date A point or period of time associated with an event in the lifecycle of the resource 29-12-2024 Contributor An entity responsible for making contributions to the resource FAJAR BAGUS W Format The file format, physical medium, or dimensions of the resource PDF Language A language of the resource ENGLISH Type The nature or genre of the resource TEXT https://repository.horizon.ac.id/files/original/9ea3a32a0db76b4cdbf060f10355d89e.pdf 26af08be1992b957be9b3998668ca92c Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Title A name given to the resource Vol 8 No 6 (2024) Text A resource consisting primarily of words for reading. Examples include books, letters, dissertations, poems, newspapers, articles, archives of mailing lists. Note that facsimiles or images of texts are still of the genre Text. Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Title A name given to the resource Utilization of HouseholdOrganic Waste intoBiogas and Integrated with IoT Subject The topic of the resource Biogas, methane, carbon dioxide, methanogenic bacteria, Internet of Things, temperature, humidity Description An account of the resource The increase in population impacts several environmental sectors, particularly the use of natural gas energy for household needs, such as LPG (Liquefied Petroleum Gas). This has resulted in the depletion of natural gas reserves and a rise in LPG imports. Additionally, the growing population contributes to the accumulation of household waste, which can lead to excessive leachate production and greenhouse gas emissions. This issue is particularly concerning in developing countries like Indonesia due to its negative environmental impact. This research aims to provide a solution and contribute to reducing household waste accumulation by utilizing organic waste to create renewable energy in the form of biogas as analternative to LPG.Biogas is produced through the fermentation of organic waste. Nutrient-rich fluids containing sugar can enhance the performance of methanogenic bacteria in biogas formation. In this study, we conducted nutritional testing on molasses and coconut water to determine which nutrients optimize biogas production efficiency by monitoring the pressure of the generated biogas. Generally, biogas comprises methane and carbon dioxide. It is important to note that excessive methane can lead to explosions, while high carbon dioxide levels contribute to greenhouse gas emissions. The quantities of methane and carbon dioxide produced during biogas generation can be influenced by temperature and humidity. Therefore, monitoring pressure, temperature, humidity, methane, and carbon dioxide levels in the biogas production process using the Internet of Things (IoT) is a prudent approach. The results indicate that a substrate mixed with molasses produces biogas at twice the pressure compared to coconutwater. Furthermore, optimal biogas production with ideal methane and carbon dioxide levelsoccurs at temperatures between 25-35°C under high humidity conditions. This suggests that mesophilic methanogenic bacteria thrive in tropical climates Creator An entity primarily responsible for making the resource Bintang Indra Maulana1, Andhika Suryamiharja2, Pradipa Catya Wisesa3, Rendy Munadi4, Sussi Sussi5 Source A related resource from which the described resource is derived https://jurnal.iaii.or.id/index.php/RESTI/article/view/5906/1000 Publisher An entity responsible for making the resource available Teknik Telekomunikasi, Fakultas Teknik Elektro, Telkom University, Bandung, Indonesia Date A point or period of time associated with an event in the lifecycle of the resource 29-12-2024 Contributor An entity responsible for making contributions to the resource FAJAR BAGUS W Format The file format, physical medium, or dimensions of the resource PDF Language A language of the resource ENGLISH Type The nature or genre of the resource TEXT https://repository.horizon.ac.id/files/original/a54579d3b98f7e084aa276cc6b9626d8.pdf faa85bb1de891f1601c9721c8540e17c Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Title A name given to the resource Vol 8 No 6 (2024) Text A resource consisting primarily of words for reading. Examples include books, letters, dissertations, poems, newspapers, articles, archives of mailing lists. Note that facsimiles or images of texts are still of the genre Text. Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Title A name given to the resource Knowledge Management Foundation and SolutionsImplementationin IndonesianGovernment HigherEducational Institution Subject The topic of the resource business intelligence; foundation;government highereducational institution; knowledge management;partial least squares-structural equation model; solutions Description An account of the resource The performance of XYZ, a Government Higher Educational Institution (GHEI) in Indonesia is assessed through two unintegrated applications. The 2023 targetperformancewas missed due to miscalculations outside applicationswhiletransforminglarge data amounts. Thus,business intelligence (BI) serves as a knowledge management (KM) toolto integratethoseapplicationsto achieveXYZ'starget. BecauseBI is costly and 70% failure rate of developmentplans, a research model was evaluated to look at currentXYZ innovation capabilityfor successful BI adoptionfrom the KM foundation and KM solutions implementation. Thisstudy useda quantitative method, employinga questionnaire for 94 civil servantsand the partial least squares-structural equation model (PLS-SEM) for data analysis. Results indicate in the KM foundation,organizational (O) negatively influences KM process application (KMP)(β = -0.292,Pv = 0.010)while KM infrastructure (I)and process (P)positively influence KMP,butKM technology (T)does not.In KM solutions,KMP is provenlinked to innovation capability whenKM systemsarelacking.Hence,severalactivitiesare suggestedto activateTthrough T, O, P, and I. The model validated80% of the hypotheses, laying the groundworkfor future studiesintowhich aspectsofT strengtheninnovation capabilities in GHEI Creator An entity primarily responsible for making the resource Boy Sandi Kristian Sihombing1, Fatoumatta Binta Jallow2, Ghina Fitriya3, Dana Indra Sensuse4, Sofian Lusa5, Damayanti Elisabeth6, Nadya Safitri Source A related resource from which the described resource is derived https://jurnal.iaii.or.id/index.php/RESTI/article/view/6005/999 Publisher An entity responsible for making the resource available Teknologi Informasi, Fakultas Ilmu Komputer, UniversitasIndonesia, Jakarta, Indonesia Date A point or period of time associated with an event in the lifecycle of the resource 29-12-2024 Contributor An entity responsible for making contributions to the resource FAJAR BAGUS W Format The file format, physical medium, or dimensions of the resource PDF Language A language of the resource ENGLISH Type The nature or genre of the resource TEXT https://repository.horizon.ac.id/files/original/5bd6938396e63175ad508487ce4b5ff3.pdf 3d7a566f615c6631793d1faa47ab0c45 Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Title A name given to the resource Vol 8 No 6 (2024) Text A resource consisting primarily of words for reading. Examples include books, letters, dissertations, poems, newspapers, articles, archives of mailing lists. Note that facsimiles or images of texts are still of the genre Text. Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Title A name given to the resource The Impact of Feature Extraction in Random Forest Classifier for Fake News Detection Subject The topic of the resource eature extraction; fake news;machine learning; Random Forest; text classification Description An account of the resource The pervasive issue of fake news spreading rapidly on online platforms.causing a concerning dissemination of misinformation. The influence of fake news has become a pressing social problem, shaping public opinionin important eventssuch as elections. This research focuseson detectingand classifyingfake newsusing the Random Forest algorithm by investigating the impact of feature extraction techniques on classification accuracy, this study specifically employs theTF-IDF method. For this purpose, we used 44,898 English-language articles from the ISOT fake news dataset. The dataset is cleaned using tokenization and stemming then split into 75% training and 25% testing. The TF-IDF vectorizer technique was applied to convert text into numeric as feature extraction. This study has implemented a Random Forest classifier to predict real and fake news. The proposed model contributes to overall classification precision by comparing it to the existing models. This fake news detectionhighlights the efficacy of the TF-IDF vectorizer and Random Forest combination whichachieved an impressive accuracy rate of 99.0%.This contribution highlights an effective strategy for combating misinformation through precise text classification Creator An entity primarily responsible for making the resource Dhani Ariatmanto1*, Anggi Muhammad Rifai2 Source A related resource from which the described resource is derived https://jurnal.iaii.or.id/index.php/RESTI/article/view/6017/988 Publisher An entity responsible for making the resource available Magister of Informatics Engineering,Universitas AMIKOM Yogyakarta, Yogyakarta, Indonesia Date A point or period of time associated with an event in the lifecycle of the resource 26-12-2024 Contributor An entity responsible for making contributions to the resource FAJAR BAGUS W Format The file format, physical medium, or dimensions of the resource PDF Language A language of the resource ENGLISH Type The nature or genre of the resource TEXT https://repository.horizon.ac.id/files/original/fe2d59175e75f260a8de08ba0813846e.pdf 06d6ea2f3362373b07ffd263a1fd5db6 Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Title A name given to the resource Vol 8 No 6 (2024) Text A resource consisting primarily of words for reading. Examples include books, letters, dissertations, poems, newspapers, articles, archives of mailing lists. Note that facsimiles or images of texts are still of the genre Text. Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Title A name given to the resource The Memory Efficiency in a Receptionist Robot's Face Recognition System Using LBPHAlgorithm Subject The topic of the resource Face recognition;LBPH;memory efficiency; Raspberry P Description An account of the resource This research aims to develop a memory-efficient face recognition system for a receptionist robot using the Local Binary Patterns Histogram (LBPH) algorithm. Given the computational limitations of the Raspberry Pi, the system utilizes optimization techniques including grayscale conversion, noise reduction, and contrast adjustment to enhance processing efficiency. Testing demonstrates that the face recognition accuracy achieves 80.5% to 85.5% in offline mode, and 72% to 81% in real-time mode, with variations due to lighting conditions and facial expressions. The robot's servo motors exhibit a response time between 1.945 and 3.561 seconds, enabling responsive and interactive user engagement. The results suggest practical benefits for deploying face recognition in resource-constrained environments, enhancing the efficiency of robotic receptionist applications. Creator An entity primarily responsible for making the resource Endang Darmawan Yudi1, Yesi Novaria Kunang2*, Ahmad Zarkasi Source A related resource from which the described resource is derived https://jurnal.iaii.or.id/index.php/RESTI/article/view/6048/987 Publisher An entity responsible for making the resource available Master of Informatics Engineering, Universitas Bina Darma, Palembang, Indonesia Date A point or period of time associated with an event in the lifecycle of the resource 25-12-2024 Contributor An entity responsible for making contributions to the resource FAJAR BAGUS W Format The file format, physical medium, or dimensions of the resource PDF Language A language of the resource ENGLISH https://repository.horizon.ac.id/files/original/25010dc52afdba70e40cbb45badbc31d.pdf fb0b6e3b8dd1728cd9c1b5152b1393e1 Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Title A name given to the resource Vol 8 No 6 (2024) Text A resource consisting primarily of words for reading. Examples include books, letters, dissertations, poems, newspapers, articles, archives of mailing lists. Note that facsimiles or images of texts are still of the genre Text. Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Title A name given to the resource Comparative Evaluation of IndoBERT, IndoBERTweet, and mBERT for Multilabel Student Feedback Classification Subject The topic of the resource BERT models;education data; finetuning; multilabel classification;sequence length;student feedback Description An account of the resource Student feedback plays a crucial role in enhancing the quality of educational programs, yet analyzing this feedback, especially in informal contexts, remains challenging. In Indonesia, where student comments often include colloquial language and vary widely in content, effective multilabel classification is essential to accurately identify the aspects of courses being critiqued. Despite the development of several BERT-based models, the effectiveness of these models for classifying informal Indonesian text remains underexplored. Here we evaluate the performance of three BERT variants—IndoBERT, IndoBERTweet, and mBERT—on the task of multilabel classification of student feedback. Our experiments investigatethe impact of different sequence lengths and truncation strategies on model performance. We find that IndoBERTweet, with a macro F1-score of 0.8462, outperforms IndoBERT (0.8243) and mBERT (0.8230) when using a sequence length of 64 tokens and truncation at the end. These findings suggest that IndoBERTweet is well-suited for handling the informal, abbreviated text common in Indonesian student feedback, providing a robust tool for educational institutions aiming for actionable insights from studentcomments. Creator An entity primarily responsible for making the resource Fatma Indriani1*, Radityo Adi Nugroho2,Mohammad Reza Faisal3, Dwi Kartini Source A related resource from which the described resource is derived https://jurnal.iaii.or.id/index.php/RESTI/article/view/6100/991 Publisher An entity responsible for making the resource available 1Department of Computer Science, Lambung Mangkurat University, Banjarmasin, Indonesia Date A point or period of time associated with an event in the lifecycle of the resource 27-12-2024 Contributor An entity responsible for making contributions to the resource FAJAR BAGUS W Format The file format, physical medium, or dimensions of the resource PDF Language A language of the resource ENGLISH Type The nature or genre of the resource TEXT https://repository.horizon.ac.id/files/original/09487b53b3fde84a924306b4e2a0e986.pdf 909128f995b2227ad61773b0f3a3b5ea Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Title A name given to the resource Vol 8 No 6 (2024) Text A resource consisting primarily of words for reading. Examples include books, letters, dissertations, poems, newspapers, articles, archives of mailing lists. Note that facsimiles or images of texts are still of the genre Text. Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Title A name given to the resource Performance and EfficiencyComparison of U-Net and Ghost U-Net in Road Crack Segmentation with Floating Point and Quantization Optimization Subject The topic of the resource U-Net;Ghost U-Net;image segmentation;memory efficiency; quantization Description An account of the resource This study presents a comprehensive comparison of U-Net and Ghost U-Net for road crack segmentation, emphasizing their performance and memory efficiency across various data representation formats, including FP32, FP16, and INT8 quantization. A dataset of 12,480images was used, with preprocessing steps such as binarization and normalization to improve segmentation accuracy. Results show that Ghost U-Net achieved a marginally higher performance, with an IoU of 0.5041 and a Dice coefficient of 0.6664, compared to U-Net’s IoU of 0.5034 and Dice coefficient of 0.6662. Ghost U-Net also demonstrated significant memory efficiency, reducing GPU usage by up to 60% in FP16 and INT8 formats. However, a sharp decline in performance was observed for Ghost U-Net in the INT8 format, where the IoU dropped to 0.2038 and the Dice coefficient to 0.3227, whereas U-Net maintained stable performance across all formats. These findings suggest that Ghost U-Net is preferable for applications prioritizing memory efficiency and inference speed, while U-Net may be better suited for tasks requiring consistent accuracy across different quantization levels. This study underscores the importance of considering bothperformance stability and memory efficiency when selecting models for deployment in real-world application Creator An entity primarily responsible for making the resource Haidhi Angkawijana Tedja1*, Onno W. Purbo Source A related resource from which the described resource is derived https://jurnal.iaii.or.id/index.php/RESTI/article/view/6089/994 Publisher An entity responsible for making the resource available Departmentof Computer Science, Informatics, Institut Teknologi Tangerang Selatan, Tangerang Selatan, Indonesia Date A point or period of time associated with an event in the lifecycle of the resource 28-12-2024 Contributor An entity responsible for making contributions to the resource FAJAR BAGUS W Format The file format, physical medium, or dimensions of the resource PDF Language A language of the resource ENGLISH Type The nature or genre of the resource TEXT