Garuda - Garba Rujukan Digital

Simulasi Aplikasi untuk Mendeteksi dan Mencegah Serangan DDoS pada Jaringan Berbasis Software Defined Network Sugiyanto Goutama; Agustinus Noertjahyana; Henry Novianus Palit
Jurnal Infra Vol 10, No 1 (2022)
Publisher : Universitas Kristen Petra

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

Software Defined Network (SDN) is one of the technological developments in computer networks. Today's computer networks generally use many network devices, where each network device has 2 functionalities, called control plane and forwarding plane. The separation of the two functions through SDN technology has the advantage of having a centralized control plane design is to make configuration and management easier. However, there is also a challenge in the form of a single point of failure that is vulnerable to Distributed Denial of Service (DDoS) attacks. Therefore, SDN requires an Intrusion Detection System (IDS) and an Intrusion Prevention System (IPS) to be able to detect and prevent DDoS attacks. This study aims to determine the level of accuracy and length of time to detect (Mean Time To Detect), as well as length of time to mitigate (Mean Time To Respond) in dealing with variations of DDoS attacks on SDN topology. This study detects in two ways, first using a signature and anomaly which will adopt the Deep Neural Network model to classify, recognize the types and patterns of DDoS attacks from a dataset with several features. The results of simulation testing with 3 types of attacks, namely ICMP Flood, SYN Flood and UDP Flood on SDN, detection with signature-IDS get MTTD and MTTR results of 7.2475 seconds and 11.74 seconds for ICMP attacks, 26.995 seconds and 11.00 seconds for SYN attacks, 20.49 seconds and 3.00 seconds on a UDP attack. While the anomaly-IDS detection does not use calculations based on MTTD and MTTR because the workings of the system can only classify per packet. So it is calculated based on the level of misclassification of the attack packet (False Negative), namely 7 packets out of 445 packets for ICMP attacks, 557 packets out of 940 packets for SYN attacks, and 2 packets out of 3120 packets for UDP attacks. Therefore, for Anomaly-IDS using the Deep Neural Network model, is still yet optimal and needs to be researched and developed further.

Aplikasi Rekomendasi Resep Menu Meal Plan Berdasarkan TDEE dan Zat Gizi Makro Pengguna Berbasis Web Dengan Pendekatan MCDA Dan EgoSimilar+ Ivana Jovita Handoko; Henry Novianus Palit; Liliana Liliana
Jurnal Infra Vol 10, No 1 (2022)
Publisher : Universitas Kristen Petra

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

Public awareness in Indonesia of the importance of a healthy lifestyle is getting higher today. Having a healthy and ideal body is the dream of people in Indonesia. One way to achieve a healthy and ideal body is to control the intake of calories, carbohydrates, protein and fat in the food consumed (TDEE and macronutrients). Everyone has different needs for TDEE and macronutrients. It is difficult to meet macro needs in accordance with the daily calorie TDEE if you do not understand how to arrange a daily food menu. TDEE and macronutrients should not be lacking nor should they be in excess, while the composition of each macro has a different calorific value for each gram unit in each food ingredient.The meal plan menu recommendation system uses EgoSimilar+ and AHPSort which is the MCDA approach, the method used in this study is used to build the "Dahar" application which is the meal plan recommendation application in this study.The results of the application and the methods used in this application are able to provide recommendations for meal plans with a tolerance of 10% calorie difference. From a scale of 1 to a scale of 5 with the meaning of a scale of 5 being the highest value, a user survey was conducted on the application and recommendation system. From the survey results, the average user satisfaction is 4.45 out of 104 respondents with the dominating scale values being scale 4 and scale 5.

Sistem Pengendali Lingkungan Greenhouse Dengan Wireless Sensor Network Untuk Mengoptimalkan Budidaya Hidroponik Kevin Hartono; Henry Novianus Palit; Anita Nathania Purbowo
Jurnal Infra Vol 10, No 1 (2022)
Publisher : Universitas Kristen Petra

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

Hydroponic cultivation is currently widely used because its various advantages. However, it is undeniable that it still has shortcomings because environmental conditions need to be engineered to resemble the conditions of the original habitat. The problem that is in this treatment process is human limitations in terms of consistencyIn this thesis, a trial will be carried out to help make it easier for users to maintenance consistently by implementing Internet of Things (IoT) technology in hydroponic gardens for monitoring and control automatically and consistently, with using sensors installed in the reservoir to get reservoir conditions and relays to regulate the flow of electricity to control environmental conditions. Furthermore, android application was developed to make it easier for users to monitor and control the hydroponic environment. So, the result is a monitoring system that can take data on several parameters, as well as controlling environmental conditions automatically by flowing the controller fluid into the reservoir, and an Android application to access measurement data that has been done previously.

Platform Big Data Analytic Berbasis Apache Spark Bagi Pemula Dalam Menyusun Data Analysis Workflow Daniel Jeremia; Henry Novianus Palit; Andre Gunawan
Jurnal Infra Vol 10, No 1 (2022)
Publisher : Universitas Kristen Petra

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

Data is a concrete foundation for decision-making. The development of technology, in turn, creates a problem in the number and complexity of data as it requires sophisticated methods to analyze. This calls for the need of big data analytics. Analyzing data quickly, simply, and robustly is now a very high requirement, especially for beginners.To combat this problem, a platform for big data analytics that is beginner-friendly is proposed in this research. This platform is created with the purpose of simplifying the process of analyzing data easily without the use of programming for beginners. Diagrams/workflows are designed to manipulate data in a drag-and-drop fashion to make it easier for beginners. Furthermore, this platform uses industry-leading technology such as Apache Spark to deal with the problems of big data analytic without being known by the user at all.A survey/demo of 12 people with 3 different backgrounds, namely commoners, beginners, and experts, is held. The obtained result indicates a positive experience in doing data analysis without programming. An average score of 4.4 out of 5 is given by the participants for how much this platform can simplify the work of data analysis. This big data analytic platform has a huge potential for beginners and professionals alike.

Penerapan Linguistic Inquiry and Word Count dan Random Forest Dalam Klasifikasi Personality Berdasarkan Data Posting Twitter Sehingga Dapat Ditentukan Gaya Belajar yang Sesuai Cristine Ferlly Wiyanto; Henry Novianus Palit; Alvin Nathaniel Tjondrowiguno
Jurnal Infra Vol 10, No 2 (2022)
Publisher : Universitas Kristen Petra

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

Big Five Personality is a powerful personality model for understanding the relationship between personality and various academic behaviors. Students' personality is very important for learning and has the potential to determine their academic achievement and learning style. However, not all students have the same knowledge, personality, and learning styles where these criteria affect learning. To find out, we usually use online tests and it takes a long time. In this study, a system was created to determine personality and learning style automatically based on Twitter post data. The method used in this research is LIWC or Linguistic Inquiry and Word Count and Random Forest. Random Forest was chosen because this method can classify class imbalances where in classifying the Big Five personalities from text data, not all of the data have the same number of personalities (extraversion, agreeableness, openness, conscientiousness, and neuroticism). The data text that will be used is data text from social media, namely Twitter with a total data of 9546 data. The results of Random Forest accuracy for balanced and imbalanced datasets are not very significant, such as the imbalanced CON personality has an accuracy of 0.499 while the balanced CON has an accuracy of 0.502 or also the imbalanced NEU personality has an accuracy of 0.502 while the balanced NEU has an accuracy of 0.519. While the results of learning style can be determined from the Big Five Personalities with an average Kendall Tau correlation value of 0.21. According to the compatibility survey of the respondents, respondents felt that the external web was more suitable with the average value of the respondent's suitability with the results of the external web of 4.5 for Big Five Personality and 4 for learning style results. Meanwhile, for the results of the program, the average obtained for the Big Five Personality is 3 while for the learning style it has an average value of 3.25

Sistem Informasi dan Rekomendasi Kegiatan Kemahasiswaan Universitas Menggunakan Content-Based Filtering pada Web App RE*ACH sebagai Pusat Informasi Kegiatan Kemahasiswaan Universitas untuk Mahasiswa Misael Rithe Setio; Henry Novianus Palit; Hans Juwiantho
Jurnal Infra Vol 10, No 2 (2022)
Publisher : Universitas Kristen Petra

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

In living the college life in Petra Christian University, each college students have his own problems. One of them is the problem of meeting the requirement for Student Activity Credit Unit (SKKK). There are also various reasons for each student to have this kind of problem. However, getting the opportunity to pass the selection as a committee member or participate in some student activities is one of the reason that the students complain most. Therefore, an information system and recommendation system for student activities must be created in a centralized platform that can be accessed by all Petra Christian University students. In helping the information system to be more precise in providing information, recommended system is added to the system so that the information related to the users can be addressed correctly to the users who really need it. In making the recommendation system, the ContentBased method with cosine similarity is used because the method tends to recommend products based to each user’s individual preferences. Users of RE*ACH application are all Petra Christian University students, so the dataset will consists of personal data from all Petra Christian University registered in the application.

Sales Forecasting pada Dealer Motor X Dengan LSTM, ARIMA dan Holt-Winters Exponential Smoothing Jennifer Soeryawinata; Henry Novianus Palit; Leo Willyanto Santoso
Jurnal Infra Vol 10, No 2 (2022)
Publisher : Universitas Kristen Petra

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

In the world of commerce, inventory is an important issue. Occasionally, motorcycle dealer X experience lost revenue due to a lack of motorcycle inventory as well as lost storage space due to under-selling motorcycles being stocked in large quantities. If the restock process is easy to do, it will answer the problem. Inventory of motorcycles at the motorcycle dealer X was sent from Jakarta to Central Sulawesi. If the motorcycle dealer X wants to do a restock, it will take a long time and expensive shipping costs. To overcome the problems at the motorcycle dealer X, a prediction or forecasting of motorcycle sales is needed. With this forecast, it is hoped that the owner of the motorcycle dealer X can determine the number and type of motorbikes that must be sent from Jakarta each month. In this study, we will use Long-Short Term Memory (LSTM), Autoregressive Integrated Moving Average (ARIMA), and HoltWinters Exponential Smoothing to forecast motorcycle sales and then compare their performance using evaluation metrics, such as the Root Mean Square Error (RMSE) and Mean Absolute Percentage Error (MAPE). From this third model, the best model for forecasting is ARIMA with the lowest RMSE (1.1339-5.8936) value for all types of motors and has the lowest MAPE values for three types of motors. If the LSTM model is compared with the HoltWinters model, the LSTM model is better at forecasting with smaller RMSE and MAPE values for most types of motors.

Analisis Sentimen Mahasiswa di Surabaya Terhadap Pelayanan Vaksinasi COVID-19 Menggunakan Beberapa Classifier Meliana Kusuma Pangkasidhi; Henry Novianus Palit; Andre Gunawan
Jurnal Infra Vol 10, No 2 (2022)
Publisher : Universitas Kristen Petra

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

Indonesia is one of the countries that are currently struggling to deal with the COVID-19 virus pandemic by providing vaccination. The government is currently trying to persuade the public to do vaccination by maximizing COVID-19 vaccination services. In reality, vaccination services still have problems with some aspects. To see various insights on vaccination services that have been implemented, therefore a research was conducted in the field of sentiment analysis to analyze public opinion. In this research, classifiers that will be used are Naïve Bayes, Support Vector Machine (SVM), Random Forest, and Light Gradient Boosting Machine (LGBM) to perform text classification and their performances will be compared with evaluation metrics. There are two types of datasets used, namely questionnaire dataset and social media dataset. The questionnaire model will be tested using a social media dataset, while the social media model will use social media dataset that will be split. The testing results show that the model trained with the social media dataset produces better performance than the questionnaire model. Of these four classifiers, the best model for aspect and sentiment classification is Random Forest

Pengaruh Feature Selection terhadap Kinerja C5.0, XGBoost, dan Random Forest dalam Mengklasifikasikan Website Phishing Michael Jonathan; Silvia Rostianingsih; Henry Novianus Palit
Jurnal Infra Vol 10, No 2 (2022)
Publisher : Universitas Kristen Petra

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

With the increase in internet users, especially websites, it provides an opportunity for phishing actors to obtain or steal personal information from users. On each website there will be a lot of information that will be used as a feature, this feature will be used to classify phishing websites. Features will be divided into 3, namely feature url, content feature, and external feature. In this study, three methods will be used, namely C5.0, XGBoost, and Random Forest. The three methods will be tested for their performance to find the best method for classifying phishing websites. In addition, this research will also utilize feature selection with the aim of removing features that have no effect so that training time can be shortened. Based on the test results obtained, it shows that C5.0 is able to provide accuracy, precision, recall, & f1-score values with an average of 93.5%, XGBoost with an average of 96.6%, and Random Forest with an average of 95.7 %. The use of feature selection in the three algorithms also shows that training time can be shortened by an average of about 3.53 times faster by using only 15 feature importance. However, with the use of feature selection, the performance on accuracy, precision, recall, & f1- score values decreased slightly even though the given decrease was not significant or had no major impact on the classification process.

Prediksi Peringkat Mingguan Lagu Pada Spotify Amerika Serikat Menggunakan Multiple Charts Dataset Dengan Berbagai Metode Christianto Imanuel Aryanto; Henry Novianus Palit; Andre Gunawan
Jurnal Infra Vol 10, No 2 (2022)
Publisher : Universitas Kristen Petra

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

In 2020, the majority of the music industry's revenue, 62.1%, came from streaming music. As a result, many music business parties are striving for a hit song, particularly on Spotify US chart. However, this is difficult to achieve because nowadays, a song's performance is determined by its performance on various music charts, not by its quality. Due to that, a study in the field of hit song science will be conducted to forecast weekly song ranking on Spotify US using data from Spotify, Shazam, Airplay, and TikTok charts. Multipler linear regression, polynomial regression, gradient boosting tree, and random forest are the methods used in this study to create models, and each model will be compared using adjusted r-squared and mean absolute error (MAE) as evaluation metrics. Random forest produced the best model, with adjusted r-squared and MAE values of 93.133% and 11.687, respectively. The usage of music attribute had a negative impact on model performance. Shazam chart, on the other hand, has been shown to have a positive impact on model performance. Meanwhile, neither the Airplay nor the TikTok charts have a definite positive or negative impact. However, both have been shown to have a very weak relation with model performance. Overall, the dataset combination of Spotify, Shazam, Airplay, and TikTok chart produced the best model in this study.

Title

Found 14 Documents
Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title Search

Found 14 Documents Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title

Found 14 Documents
Search