Computer Science (CO-SCIENCE)

Perancangan Game Role-Playing sebagai Sarana Edukasi Sejarah Menggunakan Metode Game Development Life Cycle

2023-09-27T15:53:42+07:00

Sejarah sebagai suatu kejadian yang telah terjadi di masa lampau dapat menjadi sebuah pelajaran di masa yang akan datang bagi umat manusia. Ironisnya, generasi muda jaman sekarang lebih condong untuk mengadopsi budaya global. Kemajuan teknologi yang pesat membuat pengetahuan dalam lingkup teknologi lebih diminati, namun, bukan menjadi suatu alasan bagi calon generasi masa depan untuk melupakan pondasi tempat ibu pertiwi berdiri. Salah satu fenomena yang tidak asing di dengar adalah video game. Melihat hal ini, pengembangan video game yang juga akrab di adopsi pada tujuan yang lebih serius seperti edukasi, mendorong penelitian ini, yaitu untuk merancang suatu sistem game yang memiliki nilai edukasi yang dapat menghibur dan memberikan pengetahuan kepada penggunanya tentang pesan yang ingin diberikan. Penelitian dilakukan menggunakan Metode Game Development Life Cycle yang dimulai dari tahap inisiasi untuk membentuk mindmap melalui brainstorming pengembang, pra-produksi guna membentuk low-fidelity prototype dan Game Document Design, produksi tahap dimana scene dibentuk, pengujian versi alpha oleh pengembang, hingga pengujian beta oleh pengguna potensial. Penelitian telah menghasilkan game “Keraton: Maja” yang berhasil dijalankan tanpa defect atau bug. Hasil pengujian kuesioner likert oleh beta tester mendapatkan rata-rata total nilai elemen sebesar 4,1 sehingga dapat disimpulkan bahwa sistem game dapat diterima dengan baik.

Penerapan Teknik Random Oversampling Untuk Memprediksi Ketepatan Waktu Lulus Menggunakan Algoritma Random Forest

2023-10-27T12:58:59+07:00

Punctuality of graduation is something that students yearn for, besides being important for students, punctuality of graduation is also very important for universities, this is because the aspect of student graduation is one aspect of assessment in an institutional accreditation process of a university to show its quality. One of the obstacles faced to find out whether a student can graduate on time or not is because the study period cannot be detected early, this will have an impact on late student graduation. To analyze this, a lot of research was conducted on the accuracy of student graduation, through the cumulative grade point average (GPA) obtained by students during their studies. This research on the prediction of student graduation timeliness uses a random forest algorithm model. The data used in this research object has an unbalanced number of data classes, to overcome this, a random oversampling (ROS) resampling technique is applied and also applies Split validation or division between learning data by 50% for test data and 50%. To evaluate the model built, the author uses evaluation metrics such as accuracy, recall, and precision. The results of the study showed that the proposed model can well predict compared to other models, namely with the results precision of 87.05%, accuracy test values of 90.04%, recall of 90.04%.From these results, it can be interpreted that the random forest algorithm is considered good in predicting the timeliness of a student's graduation

Analisa Peformasi Metode Rendering Website: Client Side, Server Side, dan Incremental Static Regeneration

2023-10-13T10:24:58+07:00

Website telah menjadi bagian penting dari kegiatan bisnis, pendidikan, hiburan, maupun media sosial. Optimalisasi website menjadi penting saat proses pembuatannya. Hal ini disebabakan supaya pengguna website merasa nyaman. Akses website yang cepat dan ringan menjadi salah satu daya tarik pengguna. Metode pembuatan website pada bagian Front-End diklasifikasikan menjadi 3 macam, yaitu: metode render CSR (Client Side Rendering), metode render SSR (Server Side Rendering), dan metode render ISR (Incremental Static Regeneration). Penelitian ini menggunakan aplikasi website dengan framework Next.js sebagai website untuk pengujian. Dibuat dalam website satu halaman berisi list daftar berita dari API CNNIndonesia.com. Tujuan penelitian ini untuk mencari metode mana yang optimal digunakan dalam pengembangan website. Sebagai metrik performansi digunakan kecepatan dan ukuran website. Hasil yang diperoleh adalah bahwa metode ISR sangat cocok digunakan dalam pengembangan web frontend karena lebih ringan dan lebih cepat diakses dibanding metode SSR dan CSR. Dengan rekomendasi untuk pembuatan website, jika menginginkan kecepatan berarti dapat menggunakan metode ISR untuk peramban crome jika menggunakan mozilla SSR. Jika menginginkan generate yang ringan maka menggunakan metode ISR untuk mozilla dan ISR untuk chrome.

Analisis Sentimen Ulasan Aplikasi Identitas Kependudukan Digital Pada Play Store Menggunakan Metode Naïve Bayes

2024-01-09T18:40:30+07:00

One of the government's new innovations is the implementation of Digital Population Identity which is to bring transformation in the way people prove their own identity, making it easier, safer, and more efficient in various contexts of everyday life. It is known that on October 20, 2023 the Digital Population Identity application only received 3.3 stars from 32.7 thousand reviews, this shows that the application still has not reached the maximum level of satisfaction in serving the needs of the community. Therefore, sentiment analysis of user reviews is needed to gain deeper insight into how people respond to this application, so this research aims to obtain sentiment data from the Digital Population Identity application on the Play Store in the form of reviews from users and the results of the analysis can be used as evaluation material for application developers. Data collection is done using the scrapping method through Google Colab as many as 1000 reviews whose results will be labeled negative and positive. Then the data will be cleaned and simplified through preprocessing, and will be classified with the Naïve Bayes algorithm with 90% test data and 10% training data for each iteration. The classification results are then calculated performance with confusion matrix and obtained an accuracy value of 82.23%, precission of 76.08%, and recall of 94.02%.

Optimasi Algoritma Naïve Bayes Berbasis Particle Swarm Optimization Untuk Klasifikasi Status Stunting

2024-01-05T10:27:43+07:00

Every parent wants their children to grow up healthy. Eating a healthy diet can minimize stunting. Long-term nutritional deficiencies can lead to stunting, a chronic nutritional problem that impairs physical growth and development, including low body weight and height. Preventive action against stunting is a fundamental activity that must be done immediately in the form of counseling and taking further medical action. In data mining there are several methods for extracting information including classification. There are various methods for extracting information using data mining, such as classification. In this research, researchers will apply Naïve Bayes with Particle Swarm Optimization (PSO) for the classification of stunting status in order to determine whether a child has a case of stunting or not based on gender, age, birth weight, body weight, body length, and breastfeeding. In the final results of the research, it is known that the accuracy of the truth obtained through the performance of the Naïve Bayes algorithm model is 80.69% and a score of 0.801 resulting from Area Under the Curva (AUC). Then based on the calculation results with the Naïve Bayes algorithm model with Particle Swarm Optimization, it can be obtained a truth accuracy rate of 83.06% with an Area Under the Curve (AUC) value of 0.801. Based on the final value obtained, the pattern of applying Particle Swarm Optimization to the Naïve Bayes algorithm can improve the performance of the classification method used in this research activity.

Metode Vulnerability Assesment Dalam Pengujian Kinerja Sistem Keamanan Website Points of Sales

2024-01-09T19:03:10+07:00

The development of electronic commerce through point of sales based websites is closely related to the growth rate of the internet, because electronic commerce runs through networks and Internet connections. However, the more point of sale based websites that are built, the greater the possibility of cyber attacks that could harm the website. Therefore, website security is very important to pay attention to. One method that can be used to maintain website security is to carry out a Vulnerability Assessment. Vulnerability Assessment is a process of searching for security gaps in an information system or computer network with the aim of identifying potential security vulnerabilities and taking preventative steps before an attack occurs. The vulnerability assessment technique used is using a weakness scanner application to identify security gaps in systems and applications such as Nikto, Nmap, Zenmap and Owasp ZAP. Based on testing with the Owasp ZAP tool, the results of scanning carried out on the sakupos.com website, which is a points of sales based website, show that there is a vulnerability on the website. The test results show the Level of Vulnerability (Risk Assessment) as well as recommended solutions that can be used to prevent it. There were 10 vulnerabilities detected, 7 vulnerabilities were found with a Medium risk level, 2 vulnerabilities with a Low risk level, and 1 other vulnerabilities at the Informational risk level.

Pengembangan Aplikasi Manajemen Litabmas Pendanaan Mandiri dengan Pendekatan Hierarchical Model-View-Controller (HMVC)

2024-01-16T14:17:38+07:00

In higher education, overseeing the execution of litabmas activities is a challenging and error-prone task, particularly when done manually through the physical submission of documents. Information is not always readily available, archives are frequently lost or even damaged, and tracing Litabmas external responsibility (articles, journals, patents/IPRs, etc.) from earlier periods can be challenging. This article describes how the Waterfall approach was used to construct the Litabmas management application. The Hierarchical Model-View-Controller (HMVC) technique is used to build the code architecture during the coding process. By reducing intermodular dependencies, this method makes it easier for a team to design an application. This application was developed to facilitate the management of the full cycle of Litabmas activity implementation for users, from generating Litabmas master data to initiating the Litabmas period, submitting proposals, conducting evaluations, and reporting Litabmas outputs and outcomes. The application has been tested using the black box method, and as a result, 23 use cases from six levels of users —administrator, head of institutional research, researcher, reviewer, auditor, and general visitor access rights —have been generated. All of these usecase have 100% of the expected output produced by the application.

Performance Measurement of Classification Model with Data Oversampling in Supervised Learning Algorithms for Heart Disease

2023-10-11T11:19:46+07:00

Heart disease remains a leading cause of death in Indonesia and worldwide. In the realm of data mining, class imbalance between heart disease and normal samples within datasets presents a significant challenge. This disparity can lead to model bias toward the majority class, resulting in suboptimal performance in identifying instances of heart disease. This study addresses this issue by implementing oversampling techniques, particularly Synthetic Minority Over-sampling Technique (SMOTE) and Adaptive Synthetic Sampling (ADASYN). The findings reveal that models without oversampling achieve accuracy and precision exceeding 80%, but exhibit poor class separation performance. In contrast, models employing oversampling, despite experiencing reductions in accuracy and precision, enhance their ability to distinguish between heart disease and normal classes. The top-performing model utilizing the Random forest algorithm with SMOTE attains an AUC value of 0.868, signifying a significant improvement in class separation. These discoveries provide essential guidance for the development of more effective and accurate heart disease classification models. The utilization of oversampling techniques, such as SMOTE, proves to be an effective strategy for mitigating class imbalances in heart disease data mining. While accuracy and precision may decrease, the model's capability to identify heart disease becomes more reliable, with notable outcomes assessed using AUC. This research contributes significantly to enhancing efforts in heart disease prevention and treatment through sophisticated and sustainable data mining techniques.

Analisis Sentimen Pemanfaatan Artificial Intelligence di Dunia Pendidikan Menggunakan SVM Berbasis Particle Swarm Optimization

2024-01-05T11:34:26+07:00

The utilization of Artificial Intelligence (AI) in the field of education in Indonesia has witnessed significant developments in recent years. The advancements in AI technology have opened up new opportunities to enhance the quality of education, and address various challenges faced by the Indonesian education system. This has naturally sparked diverse opinions and comments from the public, particularly on the social media platform X/Twitter. This research focuses on sentiment analysis of reviews expressed on the X/Twitter social media platform. The primary goal of this study is to develop an effective sentiment analysis method by leveraging the Support Vector Machine (SVM) algorithm optimized with Particle Swarm Optimization (PSO) for feature selection. In this research, user reviews from X/Twitter were collected and analyzed to identify positive or negative sentiments within the context of each comment. The SVM algorithm was used to classify sentiments based on similarity to comments with known sentiments. Feature Selection PSO was employed to optimize the parameters within SVM to enhance sentiment analysis accuracy. The results of sentiment analysis on comments or tweets on the X/Twitter social media platform using both SVM and PSO-based SVM algorithms indicated that the PSO-based SVM algorithm achieved a higher accuracy. The SVM algorithm with feature selection PSO produced accuracy 89.50%, precision 86.98%, recall 93.00%, and AUC 0.964. Meanwhile, the SVM algorithm had accuracy 87.50%, precision 85.46%, recall 90.50%, and AUC 0.956. This demonstrates that the use of feature selection PSO in the SVM algorithm is capable of improving the accuracy of the results.

Penerapan Metode SMOTE Untuk Mengatasi Imbalanced Data Pada Klasifikasi Ujaran Kebencian

2024-01-16T14:27:19+07:00

Hate speech is the spread of hatred towards individuals or groups on the basis of ethnicity, religion, race, and other characteristics that can lead to discrimination, violence, and social conflict. Unbalanced data can cause negative results in classification results. The Synthetic Minority Oversampling Technique (SMOTE) method is used to deal with unbalanced data. Feature extraction uses Bag of Words and TD-IDF, then the training data are oversampled using the SMOTE, SVM-SMOTE, Kmeans-SMOTE, and Borderline-SMOTE methods. This classification uses the Random Forest, Support Vector Machine, Logistic Regression, and Naive Bayes algorithms using Twitter data. The research results show that the application of the Borderline-SMOTE method to handle imbalanced data produces better performance than other SMOTE methods based on accuracy, recall,precision and F1-Score values with respective values of 84.09%, 85.25%, 84,55% and 81.16%. The Random Forest algorithm produces higher performance values than other algorithms.