JSAI (Journal Scientific and Applied Informatics)

Design and Implementation of Application Systems to Improve the Efficiency and Effectiveness of Research and Community Service Institute (LRPM) Activities

2024-05-06T00:47:44+00:00

This study aims to develop an efficient and effective information system for the Research and Community Service Institute (LRPM) using the Rapid Application Development (RAD) method. RAD was chosen for its capability to produce rapid system development with iterations that allow adaptation to changing needs. The research methodology includes planning, design, implementation, and system testing. The system development involved the use of PHP with Laravel as the framework and MySQL as the database. The results indicate that the developed LRPM information system successfully met all the predefined functional requirements, including modules for login, registration, proposal submission, proposal tracking, and data management by the admin. System testing using the black box testing method resulted in a functionality score of 100%, indicating that the system operates according to expectations and is ready for use. This system is expected to improve efficiency and effectiveness in managing research and community service activities at LRPM.

QOS ANALYSIS WITH TIPHON STANDARDIZATION ON TRIPLE PLAY SERVICES USING GPON AT PT XYZ)

2024-05-06T00:47:19+00:00

This research aims to assess the Quality of Service (QoS) in Triple Play services utilizing Gigabit Passive Optical Network (GPON) technology at PT XYZ. The research focuses on implementing the TIPHON standard (Telecommunications and Internet Protocol Harmonization Over Networks) as a framework for evaluating service quality. The research methodology involves a survey of QoS parameters, including download speed, upload speed, and voice call quality. Additionally, the study includes an evaluation of GPON network performance based on the TIPHON standard. The results indicate that the implementation of the TIPHON standard enhances QoS in Triple Play services through GPON infrastructure at PT XYZ, with a significant improvement in QoS parameters. The analysis of GPON network performance, considering the TIPHON standard, aids in identifying areas for improvement to enhance efficiency and service quality.

Improving the Quality and Performance of Underwater Image Classification using the CLAHE-CNN Method

2024-05-21T00:37:31+00:00

Research on underwater image analysis is critical because of challenges such as color distortion, low contrast, and noise in images. Various methods have been proposed to overcome this problem. To improve the quality and classification of underwater photos, this study aims to ensure improved performance using Contrast Limited Adaptive Histogram Equalization (CLAHE) and Convolutional Neural Networks (CNN) on underwater image datasets. The dataset consists of 500 JPG image with RGB channel and dimensions of 512 × 512 collected from online sources. There are classifications of sharks, eels, dolphins, sea rays, and whales in the underwater imagery dataset. The experiment was conducted using Python programming language on a computer that had 24GB RAM, Intel® Core™ i7-10510U CPU and hardware properties of Intel® HD Graphics graphics card. The results of this study show how CLAHE improved the CNN classification of underwater imagery by 0.91% in training data, 0.45% in validation data, and 2.02% in test data.

USE OF THE SYSTEM USABILITY SCALE (SUS) METHOD IN THE SIMAMURAT APPLICATION

2024-02-27T04:33:54+00:00

The application of an information system in a State University as an education provider is very important to obtain faster and more accurate information. One of the application systems used at Fatmawati Sukarno State Islamic University Bengkulu is the Mail Management Information System (SIMAMURAT). To determine the level of usefulness of a system / application, it is necessary to do testing. Testing can be done using the System Usability Scale (SUS) method. System Usability Scale (SUS) is one approach used to measure the level of usability or usability of a system or product based on user perception. This method was developed by John Brooke in 1986 and has become one of the standard methods in usability evaluation. The average score obtained from the calculation of SUS respondents using the SIMAMURAT application is 70.03. From these results, it can be seen that the Percentil Rank in the SIMAMURAT application is in the Good category, while the Grade in the SIMAMURAT application is C which shows the average, for Adjectives in this study it is classified as OK, while in Acceptability it is in the marginal category and for NPS classification, it is passive. From the results of this study, it can be concluded that the SIMAMURAT Application can be declared feasible based on Usability based on the SUS method.

Performance Analysis of Skin Cancer Multi-Class Dataset Classification Method Using KNN and HOG

2024-05-21T00:35:27+00:00

Detection of skin cancer in its early phase is a challenge even for dermatologists. This study aims to analyze the performance of classification methods on multiclass skin cancer datasets using K-nearest neighbor (KNN) and histogram of oriented gradients (HOG). The dataset is taken publicly under the name Skin Cancer MNIST dataset: HAM10000 dataset totaling 10,015 data. The first experiment used the pixels per cell parameter of 8.8 and cells per block of 2.2 to get an accuracy of 60.58%. The second experiment used the pixels per cell parameter of 8.8 and cells per block of 2.2 to get an accuracy of 60.58%. The last experiment using the pixels per cell parameter of 8.8 and cells per block of 2.2 got the best accuracy of 61.43%.

Brain Tumor Image Classification Using Gaussian Model Based on Machine Learning Based on MRI Dataset

2024-05-21T00:36:48+00:00

Early detection of brain tumors using brain magnetic resonance imaging is needed to prevent benign tumors from developing into malignant tumors. This study aims to classify brain tumors using thresholding and support vector machine (SVM) methods. The thresholding methods used in this study are global thresholding, adaptive thresholding and gaussian thresholding. The evaluation methods used are accuracy, recall, precision, and specificity. This study has used magnetic resonance imaging (MRI) based image datasets totaling 3,079 data. Overall, the accuracy of the support vector machine (SVM) algorithm and adaptive thresholding method got the best accuracy of 84.25%, while the gaussian thresholding method got 82.81% accuracy and global thresholding got 81.25% accuracy.

Analysis of IT Service Management in jd.id E-commerce Bankruptcy Using the Infrastructure Technology Information Library (ITIL v3) Framework

2024-05-06T00:46:21+00:00

E-commerce has recently become a place that is in great demand by consumers to shop. In general, e-commerce is a term used to describe the sale of goods and services via the internet. Among the many e-commerce services in Indonesia, one of them is JD.id. This research aims to look at the IT services used by JD.id previously by looking at the processes of domain service operations with the ITIL V3 framework and finding out the extent of management of the IT services used by JD.id before JD.id closed. The method used in this research is a quantitative descriptive method and the assessment of the results of this research is based on the answers to the questionnaire given to respondents using a Likert scale. The sample taken was 100 people who were JD.id users. The data sources obtained in this research are primary data and secondary data. The data analysis carried out was calculating the Maturity Level of the Service Operation subdomain. The results of the research obtained the lowest value, namely the problem management variable with a value of 3.70, meaning that the JD.id marketplace is still not very optimal in terms of preventing incidents from occurring and there is no permanent improvement process and is not very optimal in terms of continuous service improvement. However, the overall maturity level (Maturity Level) on average is at level 4 with a score of 3.97 this means that the IT service processes in the JD.id marketplace have been planned and implemented well and consistently, but automation with the information technology tools used is limited.

Identification of Spinal Disorders Based on Spine X-ray Dataset Processing Using the LBP and CNN Algorithms

2024-05-21T00:42:55+00:00

This research will use deep learning in conducting spinal x-ray image analysis but computational time problems are a problem of this study. Computations on deep learning across multiple nodes can increase training time and longer computation time compared to machine learning models. Based on experimental results, the best spine x-ray image classification results when using the CNN model with accuracy at the training stage, evaluation stage and test stage were 69.00%, 83.33% and 81.16% respectively. CNN models optimized with LBP get the lowest accuracy, with results at the training stage of 62.64%, validation stage of 75.00% and testing stage of 65.22%. LBP feature extraction turns out to have several drawbacks when combined with the CNN model, one major drawback is its inability to process global spatial information while retaining local texture information which causes LBP to be unable to capture the entire structure or context of the image, focusing only on local patterns so that many features of the image are lost. Another issue is the sensitivity of CNNs to image data, which can affect classification accuracy.

Evaluation of PLN Mobile Application User Experience Using the User Experience Questionnaire Plus (UEQ+) Method

2024-05-06T00:44:53+00:00

This research aims to explain the User Experience Questionnaire Plus (UEQ+) Method for Analyzing and Evaluating User Experience on the PLN Mobile Application. PLN Mobile is an online-based PLN application that has been downloaded by 10 million users until research was conducted from 10 May 2023 – 06 June 2023. The results of the PLN Mobile review received positive scores as well as negative reviews. Seeing positive and negative comments by application users, this prompted this research. This research is aimed at obtaining the level of user experience using scientific methods. This research will use the User Experience Questionnaire Plus (UEQ+) method which consists of 8 UEQ+ and also uses Microsoft Excel to analyze the questionnaire data obtained. As an online questionnaire, the scale that will be used in UEQ+ is in accordance with the relevant scale recommendations in the web shop product category. Based on the results of data processing from 404 respondents on each scale, the PLN Mobile application received a positive evaluation value on the scale (intuitive use, dependability, trust, content appropriateness, content quality, clarity, visual aesthetics, value) and received a mean value (1.94, 1.94, 2.03, 2.14, 2.11, 2.12, 2.01, 2.09). The results of all the important rating graph values get positive values. This shows that every scale measured in the PLN Mobile application is important.

Comparison of Activation Function Performance in the Resnet Algorithm for Rice Type Classification

2024-05-21T00:43:34+00:00

Quality checking of rice seed varieties (Oryza sativa) is an important procedure for quality assessment in the agricultural sector. The application of transfer learning algorithms has shown good results in image recognition tasks, so this algorithm is suitable for classifying rice variety images automatically. The data classes to be analyzed are Arborio, Basmati, Ipsala, Jasmine and Karacadag based on morphological, shape and color features analysis using the ResNet algorithm. The experiment used three types of models, namely ResNet-TopHat-ReLU, ResNet-TopHat-LeakyReLU and ResNet-TopHat-eLU. The ResNet-TopHat-eLU model is the best model with training accuracy of 96.61%, validation accuracy of 95.12% and testing accuracy of 78.17%.

Literature Study on Smart-Posyandu Modeling as a Platform for Monitoring and Preventing Child Stunting

2023-12-09T23:36:31+00:00

This study examines the implementation of the Smart-Posyandu platform as an innovative solution for monitoring and controlling stunting in children under five years old (toddlers) in Indonesia. The primary focus is on addressing the challenge of stunting, an indicator of developmental failure in toddlers, often caused by inadequate nutrition and repeated infections. The research findings indicate that the implementation of Smart-Posyandu, with features such as monitoring child growth, managing nutritional status, educating pregnant women on nutrition, and integrating with health services, can enhance the effectiveness of monitoring and intervening in stunting. The study also highlights the importance of collaboration among healthcare workers, the community, and the government in stunting prevention efforts. Recommendations include field implementation of the application, raising community awareness, integrating health systems, developing technology, and strategic collaboration. These findings contribute to the knowledge of applying information technology in public health and support the vision of Smart Government development in Indonesia, particularly in efforts to improve toddler health and reduce stunting.

Analisis Retinal Optical Coherence Tomography (OCT) Untuk Deteksi Kerusakan Retina Menggunakan Metode Machine Learning

2024-05-21T00:44:11+00:00

This study attempts to use support vector machine and otsu thresholding as proposed algorithm models to classify Retinal optical coherence tomography (OCT) images. In this study, there are two types implemented in classifying retinal image datasets. The first scenario is to classify using the support vector machine algorithm without the otsu thresholding method and the second scenario is to classify using the support vector machine algorithm with the otsu thresholding method with various parameter values. Based on the experimental results, classification of retina image datasets using the support vector machine algorithm without the otsu thresholding method obtained an accuracy of 63.00% while classification using the support vector machine algorithm with the otsu thresholding method with parameter values (0, 255), (50, 255), (100, 255), (150, 255) obtained an accuracy of 59.30%.

Underwater Image Segmentation with the GMM (Gaussian Mixture Model) Algorithm

2024-05-21T00:45:11+00:00

The purpose of this study was to measure the performance of Gaussian Mixture Model (GMM) technique for underwater image segmentation of seagrass objects based on datasets from autonomous surface vehicles (ASV from the Faculty of Fisheries and Marine Sciences, Bogor Agricultural University. The dataset is 640 x 480 pixel image data to support image segmentation research. There are three categories of underwater imagery: (a) underwater imagery featuring seagrass and seawater backgrounds; (b) underwater imagery featuring seagrasses, clear fish, and seawater backgrounds; and (c) underwater imagery featuring seagrasses, faint fish, and seawater backgrounds. Based on the experimental results, seagrass objects in image type (a) have almost identical colors to each pixel in the underwater image, the GMM model was able to distinguish them from the background and seawater background. The GMM model can distinguish between the background and the seawater background in image type (b), but cannot eliminate fish objects in the image. The segmentation results in image type (c) are not perfect because the GMM model removes seagrass objects that have green pixel color.

Optimizing Artificial Neural Network Performance Using Gray Level Co-occurrence Matrix (GLCM) Feature Extraction for Monitoring Elderly Movement

2024-05-21T00:34:25+00:00

Machine learning methods are used to detect elderly accidents while image processing is used to support machine learning performance so that detection performance can be better. This study used two methods used simultaneously, namely GLCM and ANN. The study consisted of preparation of human gesture datasets, preprocessing stage, application of GLCM, analysis of feature extraction results, classification using ANN and analysis of motion class detection results. Overall, the GLCM method with homogeneity parameters and ANN as a classifier obtained an accuracy of 24.32%. The GLCM method with contrast parameters and ANN as a classifier gets an accuracy of 99.84%. The GLCM method with mean parameters and ANN as the classifier gets an accuracy of 99.99%. The GLCM method with dissimilarity parameters and ANN as a classifier to classify hand movement images gets the best accuracy of 100%.

Comparison of Machine Learning Algorithm Results Based on HSV Color Model for Image Classification of Vegetable Types

2024-05-21T00:41:22+00:00

Currently, research on the classification of vegetables has made many advances. Machine learning has been proposed in recent years and has been created in image recognition, computer vision, and other fields. This study aims to classify vegetable products as part of the research of the classification of objects in charge that are inherently more complex than other subsets of object classification. This study will use the K-Nearest Neighbor (KNN) model to classify vegetable species, but with the addition of HSV color space model features. To see the performance of K-Nearest Neighbor (KNN) against other machine learning algorithms, a comparison will be made with support vector machine algorithms, logistic regression and naïve bayes. From the experimental results, the KNN algorithm got an accuracy of 80.67%, SVM got an accuracy of 72.23%, LR got an accuracy of 61.19%, NB got an accuracy of 48.77% and HSV-KNN got an accuracy of 84.33%.

Systematic Literature Review Metaverse 2019-2024, Its Scope, Potential and Challenges in the Future

2024-05-21T00:47:25+00:00

Although the metaverse is still a concept, the discussion continues to be carried out in line with the development of technology that leads to it and the latest products, such as the release of the Apple Vision Pro which is a hot topic of conversation in early 2024. Major world companies such as Apple, Facebook and Google look serious in welcoming this Metaverse. This Systematic Literature Review was carried out by extracting articles related to the Metaverse published between 2019 to 2024, which aims to determine the scope or field of development of Metaverse application related to the adoption of supporting technology, conclude what components can support the realization of the Metaverse, predict the potential and challenges that will be faced in the future, as well as submitting suggestions for future research.

Comparative Analysis of SAW (Simple Additive Weighting) Method, WP (Weight Product), and SMART (Simple Multi-Attribute Rating Technique) for Sacrificial Sheep Selection.

2024-05-06T00:43:55+00:00

In implementing urban worship, it is often difficult for shohibul qurban to determine the quality of urban animals because it has several criteria/requirements that must be met so that the animals are sacrificed according to sharia. This study aims to analyze the decision support system method for selecting qurban animals using the SAW (Simple Additive Weighting), WP (Weight Product), and SMART (Simple Multi-Attribute Rating Technique) methods. The results showed that the WP method has an accuracy rate of 99.998% so this method is the most feasible to use for the selection of sacrificial sheep when compared to the SMART and SAW methods with the calculation results in the level of suitability at 99.994% for the SAW method and 99.876% for the SMART method. Sheep 4 has the highest weight ranking of other sheep in all methods, scoring 0.923 in the SAW method, 0.1727 in the WP method, and 17.8 in the SMART method. Sheep 4 criteria is the ideal criteria for a sacrificial animal.

Identifying Land Size by Implementing Watershed Transformation Method

2024-05-31T22:46:10+00:00

This study discusses the application of the watershed transformation method to measure land area in satellite images. The main objective is to improve the accuracy of land area measurement, which is important for various applications such as agriculture, urban development, and nature conservation. Given its often time-consuming and resource-intensive nature, this study aims to develop a more efficient and faster method of processing satellite image data. The method starts with the collection and pre-processing of satellite image data to improve its quality. Next, watershed transformation is applied for image segmentation and land area measurement. The results are evaluated through accuracy analysis and comparison with reference data. From testing 20 sample data, Precision of 100%, Recall of 80%, and Accuracy of 95% were obtained. It is expected that this research can contribute to the development of a more efficient and reliable land area measurement technique.

Tomato Fruit Defect Detection System Using SUSAN Edge Detection, Statistical Feature Extraction, and CNN

2024-05-31T23:33:28+00:00

This research aims to address the problem of defect detection in tomatoes, which often compromises product quality in the agricultural industry. The difficulty in detecting defects automatically and accurately is a major challenge, so an efficient and effective method is needed. For this reason, a detection system was created by combining SUSAN edge detection method, statistical feature extraction, and Convolutional Neural Network (CNN). The SUSAN method was chosen for its reliability in detecting edges well, which is important for identifying defective areas in tomatoes. The process starts with edge detection using the SUSAN method, followed by statistical feature extraction such as mean value, standard deviation, minimum value, and maximum value of pixel intensity in tomato images. This data is then used to train the CNN model, which achieves a training accuracy of 97.50% and a test accuracy of 90%. From testing 50 tomato samples, CNN accuracy of 96%, precision of 96%, and recall of 100% were obtained. These results show that this system works well in detecting defects in tomatoes. Thus, this system is expected to improve the quality of tomato products and support the quality standards of the agricultural industry.

Application of Machine Learning to Determine the Survival of Heart Failure (Cardiovascular) Patients Using Serum Creatinine and Ejection Fraction

2024-05-31T23:34:04+00:00

Penyakit kardiovaskular menyebabkan kematian di seluruh dunia setiap tahunnya, yang sebagian besar bermanifestasi terutama sebagai serangan jantung atau gagal jantung. Gagal jantung (HF) terjadi ketika jantung tidak dapat memompa cukup darah untuk memenuhi kebutuhan tubuh. Catatan kesehatan elektronik yang tersedia dapat digunakan untuk mengukur gejala, karakteristik fisik, nilai laboratorium, dan melakukan analisis biostatistik untuk mengungkap pola dan hubungan yang tidak diketahui oleh dokter umum. Secara khusus, Machine Learning dapat memprediksi kelangsungan hidup pasien berdasarkan data dan mempersonalisasi karakteristik utama rekam medis. Artikel ini menganalisis kumpulan data 299 pasien gagal jantung dengan menerapkan algorithma machine learning menggunakan algoritma artificial neural network berbasis adaboost untuk lebih meningkatkan akurasi pada algoritma artificial neural network (ANN). Pada hasil eksperiment pada penelitian ini didapatkan akurasi algorithma artificial neural network (ANN) berbasis adaboost menjadi sangat signifikan dengan hasil akurasi menjadi 81.01%

Development of an Intelligent Violence Detection System for Bullying Monitoring Using Deep Learning Models

2024-05-31T22:46:40+00:00

Bullying in schools is a severe problem that has both short- and long-term harmful implications for victims. However, surveillance of bullying, particularly acts of violence such as kicking, pushing, and striking at school, remains inadequate. Using Artificial Intelligence is one of the recommended solutions for detecting incidents of aggression in video footage. Deep learning methods, specifically Convolutional Neural Network and Long Short-Term Memory, are used in this study to construct Artificial Intelligence for detecting acts of aggression. The model can achieve an average accuracy of up to 92%. Based on these accuracy results, the model can be implemented into online intelligent applications. It is envisaged that sophisticated software that detect such acts of aggression will be effective in monitoring bullying incidents and reducing the number of bullying cases in schools.

Digital Image Processing in Measuring Slope Using Trigonometric and Numerical Methods

2024-06-10T03:34:19+00:00

A house is a basic human need as a place of activity, shelter, and rest. One of the problems that is often overlooked in house construction is the mismatch or slope in the size of the room. Wall misalignment due to inaccurate measurements can affect building comfort and safety. This research uses technological advances in room slope measurement with trigonometric and numerical methods. The results show that the trigonometric and numerical methods have precision, recall, and accuracy rates reaching 100%. This method is effective and reliable for measuring the slope based on the captured digital image, ensuring that the measurements taken are true and accurate.

Development of Elementary School Information System Using Information Systems Implementation Research Method (ISI-RM)

2024-02-27T04:34:20+00:00

Saat ini penyampaian informasi yang dilakukan oleh Sekolah Dasar masih sangat kurang, penyampaian informasi yang digunakan hanya berupa spanduk dan informasi lisan. Dengan demikian, penyebaran informasi tentang Sekolah Dasar dinilai kurang efektif dan masih terdapat banyak kekurangan, seperti konten informasi yang terbatas dan cakupan penyebaran informasi yang tidak luas. Setelah beberapa dekade pada pengembangan penelitian sistem informasi pada bidang pembahasan tertentu berfokus pada implementasi dan penggunaan sistem informasi. Maka pada penelitian ini akan menggunakan pendekatan Information Systems Implementation Research Method (ISI-RM) dimana pada metode ini dikembangkan pendekatan metode penelitian baru dengan menggabungkan dua pendekatan kualitatif dan kuantitatif berdasarkan teori yang relevan. Hasil dari pengelitian ini yaitu pengembangan prototype sistem informasi dapat dikembangkan dengan sangat baik menggunakan pendekatan ISI-RM. Hasil ini dapat dibuktikan berdasarkan hasil pengukuran tingkat keberhasilan pengembangan prototype menggunakan teknik blackbox dengan menggunakan aspek functionality mendapatkan rata-rata nilai X lebih dari 0,5 dan mendekati 1, yaitu 0,786. Sedangkan berdasarkan aspek usability mendapatkan hasil kelayakan 94,21%.

Development of E-Commerce Information System for LPG Gas Base with Rapid Application Development (RAD) Method

2024-05-31T23:32:34+00:00

Gas is one of the fuels used by the community. The process of recording and managing transactions is still done manually, therefore the ABC 3kg LPG gas base faces challenges such as gas inventory, an inefficient ordering process, lack of information for customers, unstructured data management. To overcome this problem, the author conducts several stages starting from identifying existing problems, literature study, data collection is carried out by interviewing, observing customers, owners, analyzing needs, system design, implementation and system testing. This study aims to design and develop an e-commerce information system for the ABC 3 KG LPG gas base using the Rapid Application Development (RAD) method, because using this method can emphasize a fast and iterative development cycle, allowing continuous adjustment and improvement based on user feedback. Which has also been proven by the achievement of measurement results using blackbox techniques based on usability aspects where the results of this study reached a percentage of 97.8% success rate.

Detection of Indonesian Tropical Fruit Plant Species Using Transfer Learning Method

2024-06-28T23:16:37+00:00

This proposed research aims to determine the best algorithm performance among VGG16, ResNet and MobileNet in the process of image classification of fruit plant species. Based on the results of research experiments, the best fruit plant image classification results are the results of implementation using ResNet. The accuracy of ResNet in the training stage, evaluation stage and testing stage was 94.65%, 89.28% and 87.72% respectively. The VGG16 model obtained the lowest accuracy, with results at the training stage of 3.36%, the validation stage of 3.36% and the testing stage of 3.33%. The low accuracy of VGG16 in the classification of fruit species can be attributed to several factors. One reason is the use of weak algorithms in some cases, which limits the ability of models to accurately classify fruits. In addition, training on a small number of datasets can also contribute to lower accuracy, as models may not be able to achieve reliability.

Comparative Analysis of Machine Learning and Deep Learning Algorithms for Sentiment Analysis of Feedback Text on Lecturer Teaching Evaluation

2024-06-28T23:16:03+00:00

Evaluation of lecturer performance is very important because it helps in monitoring and ensuring that lecturers fulfill their duties effectively in maintaining integrity and teaching lecture material. By assessing lecturer performance based on criteria such as teaching, it can identify areas for improvement and provide support if needed. This study aims to determine the accuracy level of machine learning and deep learning combined with word-embedding for text analysis of lecturer teaching performance evaluation using preprocess techniques.The dataset consisted of 663 positive data, 552 negative data, and 465 neutral data. Successful in the results of the experiment, the training accuracy value for each classification method included KNN of 74.75%, SVM of 65.78%, RF of 98.58%, LSTM of 95.64% and Bi-LSTM of 95.91%. The test accuracy value for each classification method includes KNN of 59.82%, SVM of 62.88%, RF of 69.37%, LSTM of 70.81% and Bi-LSTM of 72.25%. The most superior method in processing data of 663 positive data, 552 negative data, and 465 neutral data by applying the word-embedding method, namely BiLSTM with a training accuracy of 95.91% and a testing accuracy of 72.25%.

Literature Study: Transfer Learning for COVID-19 Disease Analysis Based on Chest X-ray Dataset

2024-06-28T23:15:27+00:00

The urgency of the impact of the COVID-19 disease that attacks people around the world encourages special research, especially in the field of artificial intelligence. This study aims to conduct a literature study related to the use of artificial intelligence, especially transfer learning in analyzing COVID-19 disease based on chest X-ray datasets. The research method of this research adapts the Preferred Reporting for Systematic Reviews and Meta-Analysis (PRISMA) guidelines. The results of the analysis of this data to answer research questions regarding the transfer learning model for the analysis of COVID-19 disease based on the chest X-ray dataset, it is known that the models used are MobileNet, Inception, VGG and ResNet. MobileNetV2 can be optimized by adding a global average pooling layer, dropout layer and dense layer and get an accuracy of 98.65%. InceptionV3 can be combined with Xception and get 98.8% accuracy. VGG-16 can be combined with ResNet-50 Xception and get 98.93% accuracy. ResNet-50 can be optimized by adding a dropout layer and a dense layer and getting an accuracy of 97.65%.

Comparison of Canny Algorithm and Robert Algorithm on Edge Detection of Bengkulu Typical Batik Fabric

2024-06-28T23:17:09+00:00

This research discusses the comparative analysis of the Canny algorithm and Robert's algorithm in edge detection of Bengkulu cloth (Besurek) using the Matlab GUI interface. The selection of these two methods is to consider the balance between the quality of edge detection, noise resistance, and computational complexity. By comparing these two algorithms we can choose the algorithm that suits our needs. The purpose of this research is to make a comparison between the Canny algorithm and Robert's algorithm so as to produce the best algorithm in the edge detection process based on the number of white pixels produced. From the results of the datatest edge detection test totaling 32 images, the results of the percentage value obtained are the Canny algorithm obtaining 96.875% and the Robert algorithm 3.125%. So it can be concluded that the Canny algorithm is better in the process of edge detection based on the number of white pixels produced in each image.