Article Per Year (5 Year)
p-Index From 2020 - 2025
2.123
P-Index
This Author published in this journals
All Journal Jurnal Infra
Alvin Nathaniel Tjondrowiguno
Program Studi Informatika
Computer Science & IT

Published : 19 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 19 Documents
Search

 1   2 
Feature Selection pada Phishing Detection dengan Menggunakan Parallel Genetic Algorithm dan Ensemble Learning Alles Sandro Oktavio Gandadireja; Henry Novianus Palit; Alvin Nathaniel Tjondrowiguno
Jurnal Infra Vol 9, No 1 (2021)
Publisher : Jurnal Infra

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Phishing sites could become a threat, which retrieves personal information without the user knowing this action. Every site has numerous records, which will be converted to features. Not all features extracted are relevant. Feature selection becomes the main topic of this case. This research uses Genetic Algorithm, using Ensemble Learning as fitness function. This process requires a lot of time, parallelization then used to improve the execution time of the system. The results show that with feature selection, an improvement could be obtained. Parallelization also helps improving execution time up to 2 times faster. Using this system, it is possible to improve the effectiveness of phishing detection.
Aplikasi Pendukung Diagnosis COVID-19 Yang Menganalisis Hasil X-Ray Paru-Paru Dengan Model EfficientNet Ananta Kusuma Pangkasidhi; Henry Novianus Palit; Alvin Nathaniel Tjondrowiguno
Jurnal Infra Vol 9, No 2 (2021)
Publisher : Jurnal Infra

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

In December 2019, a new type of corona virus was detected that had symptoms of pneumonia in the seafood market in Wuhan City, Hubei Province, China. The virus then spread throughout the world, which in March 2020, WHO announced the status of the epidemic as a pandemic. WHO finally named this virus as COVID-19. COVID-19 has infected more than 105 million people worldwide, and deaths that have reached more than 2.3 million worldwide. In Indonesia alone there have been more than 1 million cases of COVID-19 and more than 30 thousand deaths in February 2021 . Based on number of cases, patient must be handled responsively. One of the supporting diagnosis for COVID-19 is Chest X-Ray. Chest X-Ray becomes one of the mandatory steps for patients to confirm and determine the treatment(s) to medicate the patients appropriately.In this study using the Deep Learning EfficientNet architecture to classify people affected by COVID-19, pneumonia, and normal from Chest X-Ray. The test results are measured by Accuracy, F1-Score, recall, precision, and specificity. With this research it is expected to be able to detect as quickly as possible so that it reduces the spread of COVID-19 and is more cost-effective because Chest X-Ray is cheaper, faster, and less radiation than CT-Scan. The result is that the accuracy in this study reaches 96 percent, and the F1-Score, Recall, Precision, Specificity is above 95 percent.
Menggunakan SPADE Algorithm Untuk Sistem Rekomendasi Film Aldy Noah; Rolly Intan; Alvin Nathaniel Tjondrowiguno
Jurnal Infra Vol 8, No 1 (2020)
Publisher : Jurnal Infra

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Nowadays, internet has become main thing in everyday life. This cause the consumption of entertainment media can be done anywhere and everywhere. One of the entertainments we enjoy is movie. With the increasing popularity of streaming media, movie fans also increase.Because of the increase of movie fans, the need of recommendation system that can recommend movie to its user also rise. System recommendation for movie is a complicated thing because of considerable amount of movie and movie fans.Sequence pattern mining is one of data mining method that can be used to gain frequent pattern from a set of data. Frequent pattern is a series of items that forms a pattern in a set of data. SPADE is one of the methods to find frequent sequence. The advantage of using SPADE is that speed in which SPADE can find frequent sequence in a data set. The benefit of using SPADE algorithm is in the speed of the algorithm to find frequent sequence. The resulting frequent sequence then can be used as a basis for recommendation to the user.
Penerapan Finite-State Machines untuk Peningkatan Performa Frame Per Second dalam Game Multiplayer Real Time Strategy Nicholas Sutikno; Djoni Haryadi Setiabudi; Alvin Nathaniel Tjondrowiguno
Jurnal Infra Vol 7, No 2 (2019)
Publisher : Jurnal Infra

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (510.603 KB)

Abstract

Games are in great demand from children to adults and games can also be played from several available platforms such as PC, mobile and console. Games have various genres such as Action, Adventure, RPG, Strategy, and many other types of genres for games. The purpose of this paper is to apply Finite-State Machines to improve FPS performance in games so that the game becomes light and comfortable to play. In previous studies FSM was used for testing AI algorithms and to assist in story formation introduce history in the game. Finite-State Machines are a method or design that will be created and implemented so that AI can make its own decisions based on predetermined conditions. Using the Finite-State Machines method because only one task is actively read so that other tasks are not read by AI in the program and light computing.The game genre that will be implemented in the game is the Real Time Strategy entitled "Attack on Toys". In the game there was a decrease in FPS performance because of the many soldiers in the game. Because this was implemented by FSM to improve the performance of the Game and the Game must still be fun when played by the player. FPS performance testing by comparing games without using FSM, using the first FSM design, and using a second design. Questionnaire testing was also conducted to find out whether the game was fun and whether the AI could work well when the game was played.Test results after the implementation of Finite-State Machines that from the results of comparison testing can increase FPS performance by up to 90%. Based on the results of the Game questionnaire it is still fun to play and AI continues to work as expected when played by players after the implementation of Finite-State Machines without reducing the quality of the game itself.
Pengenalan Intent pada Natural Language Understanding Berbahasa Indonesia dengan Menggunakan Metode Convolutional Neural Network Daniel Adi; Leo Willyanto Santoso; Alvin Nathaniel Tjondrowiguno
Jurnal Infra Vol 8, No 1 (2020)
Publisher : Jurnal Infra

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

To keep up with technological developments and people behavior, intelligent bot has become part of the business world which help them maintain good relation with their customer. Unfortunately, resource for intelligent bot in Indonesian language is very scarce compared to High Resource Language like English. Therefore further research about Natural Language Understanding in Indonesian language is needed. We use Convolutional Neural Network method to train our model. Model consist of embedding layer, convolutional layer, max pooling, flatten, dropout, and softmax layer. In the process of making model, there are many variable that can be tested such as dropout, number of filter, size of filter, etc. This research show that the amount and quality of data for each category can affect how a model understand the feature of each category which affect the overall precision. The quality of word2vec, one of the most important resource in the model can give significant impact on precision. The size of dropout can affect how the model understand the important feature of data. From various tests, we found that the best precision is 93 %. 
Prediksi Skor Pertandingan Sepak Bola menggunakan Neuroevolution of Augmenting Topologies dan Backpropagation Welly Winata; Lily Puspa Dewi; Alvin Nathaniel Tjondrowiguno
Jurnal Infra Vol 8, No 1 (2020)
Publisher : Jurnal Infra

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Football, or soccer is the most popular sport in the world. Whatmakes football special is the uncertainty and unpredictable result.There are a lot of factors that can affect the result of a footballmatch, such as strategy, skill, or even luck. Therefore, predictingthe outcome of football match can be challenging yet interestingtask.This research started with neuroevolution of augmentingtopologies, which useful to find the structur of a neural network.Then, the network produced by NEAT is optimized usingbackpropagation. Player ratings, team ratings, and playerposition are used as features of neural network.The hightest accuracies achieved are 81.5% on the final resultpredicting, and 48% on score predicting, were obtained throughNEAT network that optimized by backpropagation, with playerratings, team ratings, and total position from each sectors areused as features.However, on real life test, the player and team ratings areunknown. To calculate the player and team ratings, averagesmethods are used. Unfortunately, the network performed poorlycausing the accuracies to dropped significantly. Lack ofconsistency from player ratings are believed to be the mainproblem on calculating the player and team ratings.
Pengenalan Gambar Tempat Wisata Dengan Deep Local Feature Dan Support Vector Machine Angelika Dibijo; Agustinus Noertjahyana; Alvin Nathaniel Tjondrowiguno
Jurnal Infra Vol 8, No 1 (2020)
Publisher : Jurnal Infra

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Saving moment at a place usually done with photos. However, due to the large number of photos, organizing photos becomes difficult. A tourist may not know the name of the tourist spot he visited, do not have time to name the photo or forget the name of the place the photo was taken. Manually searching for places names through photos will take a long time.This research will conduct a trial with the implementation of the Deep Local Feature (DELF) method and Support Vector Machine (SVM) to recognize photos of tourist attractions automatically. The DELF method is an effective method for capturing image features, especially place pictures. After capturing image features, the images will be grouped based on features with SVM.The test is carried out to get the value of the parameter taking features with DELF and classification with SVM so that the recognition of tourist attractions has a high level of accuracy. For 153 image classes, DELF is performed with an image threshold of 50 and a max feature of 1000. While the classification uses SVM with kernel rbf with cost 10 and gamma 0.01. By using the DELF and SVM obtained accuracy with a test data of 0.6178.
Sistem Rekomendasi Film Menggunakan Integrated Kohonen K-Means clustering Joshua Maximillian; Henry Novianus Palit; Alvin Nathaniel Tjondrowiguno
Jurnal Infra Vol 8, No 1 (2020)
Publisher : Jurnal Infra

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

With the development of the film industry, more and more films can be watched. But because there are too many films that can be watched that cause users to be confused in finding films that match what they like. So there is a movie recommendation system to help user. The movie recommendation system itself has various ways to produce movie recommendations that users might like.The movie recommendation system using Integrated Kohonen K-Means Clustering is one of the Data Mining methods that can be used in recommending films. Intergrated Kohonen K-Means Clustering compared to Kohonen Self Organizing Maps, and also K-Means Clustering in recommending films.According to the result of Integrated Kohonen K-Means Clustering to know how many K cluster that is optimal for K-Means Clustering use the Elbow Method. To know how good the cluster you produce use Silhouette Coefficient and the score -0.389 for the Integrated Kohonen K-Means Clustering. The Mean Reciprocal Rank produced by Integrated Kohonen K-Means Clustering which score is 0.362 is better than K-Means Clustering which score is 0.003 and Kohonen Self Organizing Maps which score is 0.002.
Analisis Consumer Behaviour Pada Toko Retail Dengan Metode APRIORI-SD Nathaniel Edward; Rolly Intan; Alvin Nathaniel Tjondrowiguno
Jurnal Infra Vol 7, No 2 (2019)
Publisher : Jurnal Infra

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Retail store needs to evolve especially in digital age where ecommercebecoming more and more common and most peopleprefer the convenience of an e-commerce. One of the biggestadvantage of a “newer” e-commerce is they build they’rebusiness model on the foundation of processing data, whereasolder retail store doesn’t. Development of data mining andmachine learning are pushing older business model to do better.This journal represents the possibilities of using subgroupdiscovery as a method of analyzing transactional data. Subgroupdiscovery is a data mining technique which extracts interestingrule. APRIORI-SD is a method within subgroup discovery whereevaluation measure use by APRIORI-SD already prioritizingunusualness distribution of a given data.The result of this knowledge are able to find anomalies such asdifferentiating subgroup(s) with differences up to 50% comparedto overall distribution percentage. With the result people areable to create a better strategies in the future.
Pengenalan Gambar Botol Plastik dan Kaleng Minuman Menggunakan Metode Convolutional Neural Network Regina Valentina; Silvia Rostianingsih; Alvin Nathaniel Tjondrowiguno
Jurnal Infra Vol 8, No 1 (2020)
Publisher : Jurnal Infra

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Plastic bottles and cans are anorganic waste that cannot be decomposed by bacteria naturally and take a long time decomposed. Until now, people awareness to care about the environment is still low. Even though waste sorting is very important before the waste recycling process. According to Kediri City Environment, Hygiene and Gardening Agency, waste such as plastic and cans require special treatment for the recycling process. Until now, the process of sorting waste is still done manually by humans. This process require a lot of energy, a long time and still cannot overcome the amount of the waste nowadays.This research uses Convolutional Neural Network (CNN) method for object recognition. There have been other research about the classification of plastic waste. Both of studies only use plastic waste as the object. There is no studies yet about cans waste. Therefore, this research will carried out an introduction to plastic bottles and cans waste.Based on the result of the study, the activation function that suits the case is ELU. While using four convolutional layers, four max pooling layers, and three fully connected layers in total. This study uses 0.00001 for the learning rate, 0.8 for the dropout rate, and 50 times epoch. The result from test that were done by using this CNN model architecture is an accuracy rate of 86%.
Co-Authors Agustinus Noertjahyana Aldy Noah Alles Sandro Oktavio Gandadireja Ananta Kusuma Pangkasidhi Angelika Dibijo Ardian Teja Daniel Adi Daniel Widjojo Djoni Haryadi Setiabudi Fransisco Remon Liemena Henry Novianus Palit Ichwanto Hadi Joshua Maximillian Justinus Andjarwirawan Kartika Gunadi Kristian Halim Leo Willyanto Santoso Lily Puspa Dewi Martin Andersen Linggajaya Martina Marcelline Taslim Maximillian Christianto Nathaniel Edward Nicholas Sutikno Regina Valentina Rolly Intan Silvia Rostianingsih Welly Winata Yehezkiel Wuisang