Garuda - Garba Rujukan Digital

Penerapan Manajemen Risiko IT pada Bank X dengan Menggunakan Framework COBIT 2019 William Jordy; Leo Willyanto Santoso; Yulia Yulia
Jurnal Infra Vol 10, No 1 (2022)
Publisher : Universitas Kristen Petra

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

At Bank X, various problems occurred in business processes involving IT. problems occur such as unstable server network conditions and experiencing problems when carrying out business processes inputting data. The purpose of this thesis is to find out what factors or causes are the most influential in the use of IT in Bank X's business processes and provide a response to existing risks based on the 2019 CobiT guidelines with the Align, Plan and Organize (APO) domain in the APO 11 Managed Quality process. and APO 12 Managed Risk. Methodology The research will be conducted by examining the capability level and conducting a risk assessment using the OWASP standard in the APO11 and APO12 domains in accordance with the results in the Mapping Alignment Goal (AG) and Enterprise Goal (EG) as well as the BSC dimensions. Based on the results of research conducted, the authors found several risks that have an impact on the company's IT business processes along with the responses and solutions provided. The solution given is to Mitigate or Avoid depending on the risk severity of the risks. The conclusion of this research is that the IT Division has an important role in running the company's business processes. In addition to acting as support, the IT division also has a role in software development in customer banking applications, respondent transparency and the IT division as the company's support system are important factors in assisting this research.

Sistem Pakar Diagnosa Penyakit Sistem Saraf Pusat dengan Metode Backward Chaining dan Certainty Factor Felix Felix; Leo Willyanto Santoso
Jurnal Infra Vol 10, No 1 (2022)
Publisher : Universitas Kristen Petra

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

The nervous system or system is one of the smallest parts of the organs in the body, but its the most complex part. The Central Nervous System, namely brain (encephalon) and spinal cord, is the center of integration and control of all body activities so that will be very dangerous if our nervous system has problems, given that death can occur due to nervous system problems. Minimal knowledge and information make it impossible to know which disease that the nervous system is suffering from. Therefore, we need an expert who is skilled on nervous system diseases and their prevention.Based on the facts above, this final project can help us to diagnose central nervous system diseases and anticipate if someone have the disease. This application is made based on web base (PHP) and using XAMPP as MYSQL database server. In this expert system the user will choose the disease they want to diagnose first. Then, the user answers several questions related to the existing symptoms. After the user answers all the questions, the results of the diagnosis will appear along with the level of confidence and solutions that can help anticipate the disease.

Deteksi Masker Wajah dengan Metode Convolutional Neural Network Ivan Hartono; Agustinus Noertjahyana; Leo Willyanto Santoso
Jurnal Infra Vol 10, No 1 (2022)
Publisher : Universitas Kristen Petra

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

The computer must be able to recognize the area which is a face object in the image in order to facilitate the detection of face masks used by humans. Deep learning is artificial intelligence with simple representations that have hidden layers to process data that can build complex concepts. Deep learning can be trained to detect an object and classify objects. There are many deep learning algorithms that can be used for the model recognition process, for example for object classification using MobileNet, VGGNet, DenseNet, GoogLeNet, AlexNet, and others while for object detection you can use You Only Look Once, SSD Resnet, Multi-task Cascaded Convolutional Neural Network (MTCNN), HyperFace and others. The object detection system can use two combinations of algorithms, namely the object classification method and the object detection method. The method for recognizing mask objects on human faces is CNN (Convolutional Neural Network).The CNN method is the development of the Multilayer Perceptron which is designed to process two-dimensional data. CNN method is very good in processing spatial data and classifying objects [1]. After training the model with VGGNet, the next method is to detect an object using the SSD ResNet module.

Implementasi Text Summarization pada Review Aplikasi Super di Google Play Store Menggunakan Metode Maximum Marginal Relevance Dion Alexander Louis; Silvia Rostianingsih; Leo Willyanto Santoso
Jurnal Infra Vol 10, No 2 (2022)
Publisher : Universitas Kristen Petra

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

Super App is an app for reseller agents who sell and distribute basic necessities in tier 2, 3 cities and rural Indonesia. The Super app has been downloaded by around 50,000 users on the Play Store. Various reviews or reviews have also been given by users who have downloaded the Super application. Whether we realize it or not, customer opinions / reviews given on Google play, a little or a lot, will have an influence on potential customers. Based on the problems that occur, this research will implement a text summarization program on Super App reviews with the implementation of the MMR and TF-IDF methods, so that from the large number of existing reviews, only a few important sentences can be extracted, so that the conclusion making process will become easier. The results of the research using the MMR method produced an average precision value of 40.4% in 3 trials, and with the highest precision value of 60.4% in the experiment using the parameter value = 0.7

Sistem Pendukung Keputusan Pemberian Kredit berdasarkan Klasifikasi Kelancaran Pembayaran Kredit Menggunakan Metode VIKOR pada Bank XYZ Daniel Hartono; Leo Willyanto Santoso; Silvia Rostianingsih
Jurnal Infra Vol 10, No 2 (2022)
Publisher : Universitas Kristen Petra

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

Banks must carry out complex assessments before being able to determine who is the most eligible prospective debtor who can be given a loan. This is due to limited funds and the risk of bad credit cases. The limited manpower and manual processes cause the whole process of lending at XYZ Bank to be prone to human error and become inefficient. As a solution for XYZ Bank to overcome existing problems, a credit decision support system is needed that can assist XYZ Bank in selecting and determining prospective debtors who can be given loans. Therefore, in this study, the KNearest Neighbor method was used to assist XYZ Bank in predicting the smoothness of credit payments of a prospective debtor. Then, this research continues with ranking using the VIKOR method to determine who is the most ideal debtor candidate to be given a loan. Based on the results of the classification test using both training data and new data, the highest accuracy is obtained at 100% for each type of loan. Based on the results of the ranking test, the accuracy of the business loans ranking is 83.33%, the accuracy of the consumer loans ranking is 80.33%, and the accuracy of the various-purpose loans ranking is 70%. The results of the questionnaire evaluation in system testing conducted by 6 respondents assessed that the application design was 76.67% good, the application functionality was 86.67% good, the ease of use of the application was 83.33% good, the application answered the needs was 86.67% good, and the overall application was 90% good.

Deteksi Rompi dan Helm Keselamatan Menggunakan Metode YOLO dan CNN Rescky Marthen Mailoa; Leo Willyanto Santoso
Jurnal Infra Vol 10, No 2 (2022)
Publisher : Universitas Kristen Petra

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

Construction workers face a high risk of injury and are prone to accidents while performing their duties. Several factors increase the chance of accidents, including open and heated environmental conditions, building heights, sharp objects, and others. The use of personal protective equipment (PPE) is essential to anticipate or reduce the risk of accidents that may occur. However, it is not uncommon for construction workers to forget or purposefully disregard personal protective equipment. To address these problems, a system capable of detecting personal protective equipment for construction workers is required. This study used the You Only Look Once (YOLO) method to detect the head and body parts of the inputted image. The detected body parts were then cut and processed using the Convolutional Neural Network (CNN) method with the ResNet50 model for classification. The training process with the ResNet50 model was modified on hyperparameters including learning rate, epoch, dense layer, dropout layer, data augmentation, and freeze layer to compare its performance with the model before modification. The results showed that the YOLO model has a very high level of detection speed with good accuracy. Meanwhile, the modified CNN model performed well with an average accuracy value of 96%

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.

Sistem Pakar Diagnosa Penyakit pada Anjing Menggunakan Metode Forward Chaining dan Certainty Factor Kevin Shaquille Limanuel; Leo Willyanto Santoso; Silvia Rostianingsih
Jurnal Infra Vol 10, No 2 (2022)
Publisher : Universitas Kristen Petra

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

Dogs' diseases have different treatments which affect the owner on how to treat them. If the treatment given is not proper, even a minor disease could be fatal, which will be very detrimental to the dog owner and also the dog itself. The problem that the author wants to address is by utilizing a website that functions to diagnose common dog disease by using an expert system based on the forward chaining method and the certainty factor method to diagnose if there are any symptoms in dogs. Tests were also carried out on a collection of interview data and also from the expert in the form of disease symptoms and the program that was made are able to diagnose dog disease with the results of the method test being able to achieve an accuracy value of 80%.

Aplikasi Marketplace Vendor Lamaran dan Pernikahan berbasis Android Yansen Tri Utomo; Silvia Rostianingsih; 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 marriage or application, it is something that takes quite a long time, especially in choosing vendors, venues, and WO. With the Mywedding mobile application, it is made to assist in choosing any vendors that can be used for their wedding or application as well as provide the best recommendations according to the criteria of everyone who will prepare for a wedding. Content based filtering method used to collect data and assess suitable vendors to be recommended by the system to users. Based on the results of the tests carried out, the content-based filtering method succeeded in providing recommendations according to the wishes of the user so that they can find out which vendor can be used, although sometimes the user profile needed is more specific to be more accurate. which can be used for their wedding or proposal and provide the best recommendations that match the criteria of everyone who will prepare for a wedding. Content based filtering method used to collect data and assess suitable vendors to be recommended by the system to users.

Penerapan Metode KNN-Regresi dan Multiplicative Decomposition untuk Prediksi Data Penjualan pada Supermarket X Calvin Christopher Kurniawan; Silvia Rostianingsih; Leo Willyanto Santoso
Jurnal Infra Vol 10, No 2 (2022)
Publisher : Universitas Kristen Petra

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

Supermarket X is one of the supermarkets in West Nusa Tenggara that needs a way to predict sales in the future. This prediction is needed by Supermarket X to estimate the purchase plan because so far there have been frequent stockouts or oversupply which have caused losses to the company. Based on the problems that occur, this study applies the KNN Regression and Multiplicative Decomposition methods in predicting Supermarket X sales so that supermarket managers can design a strategy to make sales in the future. The results show that predictions based on divisions, departments, categories, sub categories, and products have a smaller average error rate when using the Multiplicative Decomposition method with RMSE = 492.89 and MAPE = 0.29, while the KNN Regression method has RMSE= 757.77 and MAPE= 0.36

Title

Found 13 Documents
Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title Search

Found 13 Documents Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title

Found 13 Documents
Search