I Gede Santi Astawa’s scientific contributions

Mathematics is a very important science because it is related to human life. Therefore, mastery of mathematics is absolutely necessary and mathematical concepts must be understood correctly. In Indonesia, students' ability to understand mathematics is still relatively low. From this, understanding of mathematical concepts needs to be improved again so that students clearly understand mathematical concepts. In order to improve understanding of mathematical concepts, various efforts were made. One of the efforts carried out by the author is a semantic web-based learning media system using concept maps in basic mathematical ontologies. This system will use ontology to build a system that will represent a mathematical concept as a domain of knowledge explicitly about other concepts related to previous concepts by giving meaning, properties, and relations so as to form a knowledge base. In this research, the writer uses methodological method to build the basic mathematical ontology and the system development will be done by prototyping method. The system built has 2 features, namely exploration and semantic search, with the aim that the existing knowledge in the system can be accessed systematically and according to user needs. To ensure that the system built can run as expected, system testing with black box testing has shown which features in the program run according to their functions and ontology evaluation using OntoQA on Schema Metrics carried out by measuring Relationship Richness (RR), Attribute Richness (AR), and Inheritance Richness (IR). In the evaluation of this ontology, it shows that the Relationship Richness (RR) value is 0.06; Attribute Richness (AR) value is 0.5; and the value of Inheritance Richness (IR) is 14.5.

The knapsack problem is problem that is still often found in everyday life, one of which is the problem of selecting goods to be transported into containers for delivery of goods. This knapsack problem can be solved by using various optimization algorithms, one of which is the genetic algorithm. This study aims to design a genetic algorithm model to solve the 2-dimensional knapsack problem. 2-dimensional knapsack problem is a knapsack problem that has 2 constraints and in this study, the constraints used were weight and volume.. The evaluation results of the genetic algorithm will be compared with dynamic programming. From the evaluation results that have been carried out, it can be concluded that genetic algorithms can produce near-optimal results with faster computational times than dynamic programming.

Rote ndao Ikat Weaving has interesting characteristics in each fabric motif because it has different motifs which indicate the ethnic differences contained in each of the resulting motifs. Rote Ndao weaving has a variety of motifs that are still unknown to many people, so in this study a classification of motifs of rote ndao woven fabrics was carried out using the Convolutional Neural Network method. Weaving motif classification uses 3 motifs with a total of 1050 data including 70 data for Ai Bunak, Dula Kakaik and Lafa Langgak motifs each. Data for 3 fabric motifs is divided into 80% training data and 20% testing data. The classification of the Rote Ndao Weaving motif is carried out by building an architectural model of MobileNetV2 plus Dropout and using a Learning Rate of 0.0003 which is trained and evaluated using K-Fold Cross Validation with a value of K=5, obtaining an accuracy of 93%; The precision is 94% for the Ai Bunak motif, 88% for the Dula Kakaik motif, and 100% for the Lafa Langgak motif. Recall of 87% for the Ai Bunak motif, 94% for the Dula Kakaik motif and 100% for the Lafa Langgak motif. Then the F-Score value obtained is 90% for the Ai Bunak motif, 91% for the Dula Kakaik motif and 100% for the Lafa Langgak motif.

The closest route optimization problem is a classic problem that is still often found in everyday life, one of them is the Traveling Salesman Problem (TSP). This study aims to find solutions in the form of optimal distances to various tourist destinations in Banyuwangi. In this study, the author uses the Python programming language as the backend and the Kotlin programming language as the frontend of the Android application as part of Modern Android Development. The results of the system evaluation, namely Black Box and User Acceptance Test on the application, show that the system that has been carried out can be concluded that the functional system can work well without experiencing bugs or problems on every available feature and page. In designing the application, the author uses the Modern Android Development pattern.

Banyak kalangan yang memiliki modal saat ini beramai-ramai memborong saham di stock market dengan harapan harga saham tersebut akan naik saat pandemi Covid-19 berakhir. Sebagai seseorang yang ingin mencoba berinvestasi di stock market harus mampu memperkirakan untung dan rugi dari pembelian saham. Salah satu cara yang dapat membantu pertimbangan dalam pegambilan keputusan membeli dan menjual saham adalah melakukan prediksi. Terdapat banyak algoritma yang dapat digunakan dalam prediksi salah satunya adalah metode Adaptive Neuro Fuzzy Inference System (ANFIS) yang merupakan penggabungan dari algoritma Logika Fuzzy dan Jaringan Syaraf Tiruan. Pengaplikasian metode ANFIS memerlukan struktur ANFIS yang baik dengan pemilihan jumlah dan tipe membership function yang tepat. Pada penelitian ini membership function tipe gaussian dan gbell digunakan karena memiliki kelebihan yaitu memungkinkan perubahan halus dan dapat mengakomodasi ketidaktepatan dalam pengukuran sehingga cocok dengan pola data histori yang bergerak secara halus di satu waktu. Pada penelitian ini diperoleh bahwa tipe gaussian memiliki akurasi yang lebih baik dibandingkan tipe gbell sebesar 97.87% untuk memprediksi harga tertinggi saham Tencent Holdings Limited dan tipe gbell memiliki akurasi yang lebih baik dibandingkan tipe gaussian sebesar 97,8% untuk memprediksi harga tertinggi saham Take-Two Interactive.

Historical documents in the form of ancient manuscripts are one form of the Indonesian nation's cultural heritage that deserves to be important, one of which is Lontar. Currently not many people can read the writings in palm leaves, therefore, ancient manuscript collectors have made efforts to digitize ancient manuscripts. The digitization of ancient manuscripts aims to improve the image quality of ancient manuscripts with the help of computers. Digitization requires an image quality improvement process by performing noise reduction and edge detection and line segmentation on digital images of ancient manuscripts. In this study, the noise reduction process uses the Mean Filter method, edge detection uses the Sobel operator, and line segmentation in this study uses the A * Path Planning Algorithm. Based on research conducted on 24 lontar images, line segmentation process obtained an accuracy of 95%.

Penyakit liver disebabkan oleh peradangan pada hati. WHO menunjukkan hampir 1,2 juta orang di asia tenggara dan afrika per tahun meninggal akibat penyakit ini. Oleh karena itu diperlukan diagnosis sesegera mungkin untuk mendapatkan penanganan lebih lanjut. Untuk melakukan diagnosis diperlukan algoritma klasifikasi di mana pada penelitian ini menggunakan algoritma C4.5. Namun algoritma tersebut kurang optimal untuk membentuk suatu pohon keputusan karena mengharuskan memuat seluruh kasus ke dalam memori. Maka dari itu diperlukan optimasi menggunakan algoritma genetika untuk membentuk rules yang lebih sederhana dengan cara menyeleksi atribut dan mencoba berbagai kemungkinan kombinasi atribut sampai didapatkan kombinasi paling optimal. Pada hasil evaluasi, rules yang dihasilkan dengan optimasi menjadi lebih sederhana yaitu sebanyak 52 rules jika dibandingkan tanpa optimasi yang lebih kompleks yaitu sebanyak 145 rules. Lalu pada evaluasi akurasi, rules dengan optimasi menghasilkan akurasi yang lebih baik yaitu sebesar 70,7% jika dibandingkan dengan akurasi rules tanpa optimasi sebesar 68,9%.

Many universities have started using e-learning as a medium of learning and teaching. However, the system on e-learning is still vulnerable to attacks such as identity theft and authentication on e-learning systems. One of the way to secure data on a network is implement a Virtual Private Network (VPN) on the network which can make a network private. That way, the security on the port on the E-learning server will be guaranteed. But attacks on servers more often occur on port 23 (telnet) because telnet does not encrypt the connection process. One of the attacks that is often used is a brute force attack. Based on these problems, research has been carried out to secure port 23 (telnet) and port 80 (e-learning website) on the e-learning server from brute force attacks by implementing the port knocking method. Port 23 and port 80 on the e-learning server that uses the port knocking method are always closed so that the client must do port knocking before being able to use port 23 and port 80 on the e-learning server. By implementing port knocking on the e-learning server, brute force attacks can be avoided because the attacked port is closed.

Education is an important thing in life. In practice, education consists of teachers and students. A teacher teaches a subject to students. Students are taught by several or more teachers. There is a provision in which a class of students can only receive one subject at a time. Teachers can also only teach one subject at a time. This requires a system that is used to regulate so that these provisions can be fulfilled without ignoring other provisions. The system used to manage these problems is a subject scheduling system. This system regulates the class placement of students with the subjects they study. In manual implementation, the process is very inefficient in terms of time, human resources and thought power. The method that can be used to solve this problem is the genetic algorithm. A genetic algorithm is a heuristic method or procedure inspired by the natural selection process. Genetic algorithms are generally used to produce high-quality solutions to problems in the form of optimization and search by relying on biologically inspired operators such as mutation, crossbreeding, and selection

Indonesia memiliki budaya yang beragam di setiap daerahnya. Salah satu contoh di Bali, terdapat budaya yang disebut tembang atau lagu tradisional Bali. Generasi muda saat ini kurang memahami informasi lagu tradisional Bali, hal tersebut disebabkan karena kurangnya informasi lagu tradisional Bali yang terkumpul secara eksplisit. Pada penelitian ini, dilakukan pengembangan ontologi untuk mendokumentasikan informasi lagu tradisional Bali dengan menggunakan metode Methontology. Ontologi kemudian diimplementasikan ke dalam sistem manajemen informasi lagu tradisional Bali yang dibangun menggunakan metode Prototyping. Sistem yang dibangun memiliki fitur pencarian semantik dan penjelajahan semantik yang bertujuan agar informasi mengenai lagu tradisional Bali dapat dikumpulkan dan diakses secara sistematis dan relevan. Dalam memastikan fungsionalitas dan pemahaman pengguna terhadap sistem, dilakukan pengujian Black-Box yang melibatkan 30 orang peserta dengan memberikan tugas penjelajahan dan tugas pencarian. Hasil yang didapatkan adalah sistem yang memiliki fungsionalitas baik, serta rata-rata partisipan dapat menjawab semua tugas dengan benar. Dari segi persepsi kegunaan dan kemudahan penggunaan sistem didapatkan hasil analisis data kuesioner menunjukkan rata-rata peserta setuju bahwa sistem yang dibangun sangat berguna dan mudah untuk digunakan.