cover
Article Per Year (5 Year)
All
Home Page
OAI Link
Editorial Team
Contact
Reviewer
Google Scholar
Contact Name
-
Contact Email
-
Phone
-
Journal Mail Official
-
Editorial Address
-
Location
Kota surabaya,
Jawa timur
INDONESIA
Jurnal Infra
Published by Universitas Kristen Petra Surabaya
ISSN : -     EISSN : -     DOI : -
Core Subject : Science,
Computer Science & IT
Arjuna Subject : -
Articles 33 Documents
 1   2   3   4 
Search results for , issue "Vol 9, No 1 (2021)" : 33 Documents clear
Deteksi Aktivitas Manusia Berdasarkan Data Skeleton dengan Menggunakan Modifikasi VGG16 Daniel Subroto; Liliana Liliana
Jurnal Infra Vol 9, No 1 (2021)
Publisher : Jurnal Infra

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

Abstract

In general, detection of human activity is carried out to detect activity in daily life. With further development, the detection of human activity is utilized to detect suspected activity (not routine) as an early warning application. Detection of human activity in early warning applications will then be implemented in other systems. However, there are several problems in detecting human activity, among others, the presence of variations in performing an activity, the movement of transitions between activities, and the similarity of movements in different activities. Detection of human activities with existing variations can be done if utilizing a deep learning approach to conduct the training process. The deep learning method used is VGG16. VGG16 will receive input in the form of skeleton data images. The skeleton data used is obtained from the NTU RGB+D dataset. Skeleton data will be represented as 2D images by going through a process of covering the crop, converted into grayscale, resizing, and connecting 10 images into 10 channels for each sequence of activity sequences. To detect human activity is applied transfer learning on VGG16 that is changing the fully connected layer. VGG16 modification test results with skeleton data representation resulting in the highest accuracy rate of 54.59%. This level of accuracy is obtained from model testing using the same dataset as the training dataset. The VGG16 modification is still the best model based on testing with other Convolutional Neural Network models. Modification of VGG16 can classify indoor activities.
Sistem Monitoring Penggunaan Air di Apartemen/Mal MultiTenant dengan NodeMCU dan Raspberry Pi Albert Ong; Andreas Handojo; Resmana Lim
Jurnal Infra Vol 9, No 1 (2021)
Publisher : Jurnal Infra

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

Abstract

As survey results conducted, it was found that it was difficult tomonitor water consumption and to manage water consumption.Water consumption become very unregulated and the surveyconducted found that respondents agreed with the existence of aweb-based water consumption monitoring and managementsystem. In this study, a trial will be conducted to help solve thisproblem by modernizing water meters or implementing Internetof Things technology on each water meter. Then for paymentsystem will be using a token system. The implementation of IoT inwater meters is in terms of measuring water consumption andlimiting the water consumption. The results of the system foundthat the system was capable of monitoring water consumption,limiting water consumption, and increase water consumptionlimit using token.
Penerapan Procedural Content Generation untuk generasi level dalam game Mythical Maze Hansen Krisna Putra; Liliana Liliana
Jurnal Infra Vol 9, No 1 (2021)
Publisher : Jurnal Infra

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

Abstract

Level design is one of the issues in creating a multi-level game application. In level design, some of  the important keypoints are the variation of the level, the time to create it, and the locations of objects inside of a level, for example the player’s starting point and the exit point. Procedural Generation can be used to simplify the process of designing a level.In the implementation of Procedural Generation, the first step is to determine the size of the desired level or terrains. The size of a terrain will affect the complexity of a level, where it will affect the amount of rooms and corridors that is made. Next, rooms with various sizes are made and placed within the level with a desired pattern, which is connected by placing corridors that acts as a connector between two rooms. The last step to create a level is to place the objects inside the level, such as player character and the end point, where a distance is given between two objectsFrom the methods that are used, a level can be automatically made with multiple variations. The levels themselves are well made in which all of the rooms are connected by corridors and the player can reach the exit point from any given poinit in a level. This method always results in a terrain where there is a line or route between the starting and exit points. Each level’s difficulty is unique, that is affected by complexity, the distance between objects, and the time given to finish a level. The player doesn’t always able to finish a big level. This method can be used to produce variants of a maze, and to add the difficulty, the terrain itself can be enlargen to add complexity
Analisis Strategi Breakout dalam Pengambilan Keputusan Order Pada Trading Forex Vincentius Bintang Febrianto; Agustinus Noertjahyana
Jurnal Infra Vol 9, No 1 (2021)
Publisher : Jurnal Infra

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

Abstract

Trading on forex with a strategy is an effort made by traders so that trading can be more profitable, one of which is using a breakout strategy and cannot be separated from determining the range of this strategy. However, in trading using a breakout strategy, sometimes some traders are not precise in determining the range. This can cause traders to lose profits. Therefore, it is necessary to establish an expert advisor program that is able to help traders make decisions in determining ranges. The expert advisor program mechanism that is created is to determine the initial reference range first to become the range amount that will be used for trading using the breakout strategy. The initial reference range is then copied to the highest and lowest prices in history. The program is then tested through backtesting.The results of tests carried out on the EUR / USD, GBP / USD, and USD / JPY pairs for 3 years can generate profits from obtained 4 initial reference ranges in various conditions of price movements. Of the three ranges of each pair on the H1 time frame, the most profitable range is the 107 pip range in the EUR / USD pair with a profit of 2528.64, the 100 pip range in the GBP / USD pair with a profit of 2628.31, the 74 pip range in the USD / JPY pair with a profit of 1691.47. In the H4 time frame, the most profitable range is 90 pips in the EUR / USD pair with a profit of 2093.06, 99 pips in the GBP / USD pair with a profit of 2429.07, 100 pips in the USD / JPY pair with a profit of 1279.94.
Dynamics Difficulty Adjusment Metode Evolutionary MCTS with Flexible Search Horizon pada Multi-Action Adversarial Games untuk Penyesuaian Tingkat Permainan Andhika Evantia Irawan; Liliana Liliana; Hans Juwiantho
Jurnal Infra Vol 9, No 1 (2021)
Publisher : Jurnal Infra

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

Abstract

Dynamic Difficulty Adjustment (DDA) is a method that modifies AI behavior to suit the player's abilities. So far, research on DDA in Monte Carlo Tree Search has been able to provide an appropriate level of challenge. However, the advantages of MCTS in finding solutions to long-term strategies have not been maximally implemented because so far it is only used in 2D real-time fighting games, which are short-term strategy game.This study combines DDA with evolutionary monte carlo tree search with flexible horizon (FH-EMCTS). FH-EMCTS is combination of vanilla MCTS with an Evolutionary algorithm. This method increases the length of the search space to certain extent. Giving DDA to FH-EMCTS is done by changing the way of selecting actions and assessing each node.The result of this research is that AI agents that use FH-EMCTS with DDA can be implemented into multi-action adversarial game and can provide balanced level of difficulty to other AI agents and humans. Based on the results of survey of AI agents against humans, it shows that the most fun and realistic AI agents are not the AI agents who have the best ability of winning percentage but AI agents who have win rate of around 50%.
Game Real-Time Strategy Conquest War Berbasis Web Menggunakan Finite-State Machine Jimmy Christian Tanian; Alexander Setiawan; Lily Puspa Dewi
Jurnal Infra Vol 9, No 1 (2021)
Publisher : Jurnal Infra

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

Abstract

With the development of technology today, games are one of the activities that are popular for many people. Starting from children and even adults are also playing videogames. strategy game is the one of them. Of the many real-time strategy games available, there are still many real-time strategy games today that are too complex and difficult for players to understand. In this thesis, a real-time strategy game will be made that is easy to understand and play but does not eliminate the strategy element in the game.In order for this game to be even more interesting, Artificial Intelligence (AI) finite state machine (FSM) was added to this game to regulate the opponent's movements in order to make the game lively. In this game, there are features i.e. save/load to save and load data and sound settings to adjust the sound. The result from 5 respondents that give rating : 76% game control, 68% game graphic, 80% game sound , 80% gameplay dan 80% tutorial clearance.
Perbandingan dan Analisis Metode Artificial Neural Network dan SIRD pada Kasus Covid-19 di Surabaya Juan Felix Nyoto Santoso; Alexander Setiawan; Silvia Rostianingsih
Jurnal Infra Vol 9, No 1 (2021)
Publisher : Jurnal Infra

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

Abstract

On March 2, 2020, President Joko Widodo's announcement regarding the COVID-19 virus has made its way to all of Indonesia, serving as a warning to the people. The virus continues to grow and spread its influence across cities, one of which is Surabaya. Surabaya attained the 'crimson zone' status on 2nd of June, 2020 due to the drastic increase of positive COVID-19 cases which tallies to 2748 people. The rapid pace at which COVID-19 spreads results in a high death rate.This research was done to try and prevent high casualty rates by predicting the need for health equipment, isolation rooms, medical personnel, and the need for personal protective equipment (PPE) for COVID-19 patients. There were two methods used for the sake of predicting, namely the Artificial Neural Network (ANN) and Susceptible Infectious Recovered Decease (SIRD) methods. The methods in question will have their accuracies tested using error measurement methods which include the Mean Absolute Deviation (MAD), Mean Square Error (MSE), and Mean Absolute Percentage Error (MAPE). After measurements have been made, the prediction results from these 2 methods will be utilized to calculate the needs for equipment, isolation rooms, medical personnel, and PPE needs based on the regulatory patterns owned by the S, N, and X hospitals. Based on the results of the website implementation analysis, the ANN method is shown to have average error rates of 53.1733 for training and 89.73 for testing based on the MAD method, 6581.09 for training and 22953.9067 for testing based on the MSE method, and 17.7367% for training and 16.3067% for testing based on the MAPE method. The SIRD method is shown to have average error rates of 309.81, 150496.08, and 30.2% for the MAD, MSE, and MAPE methods respectively.
Penerapan Algoritma TextRank dan Dice Similarity Untuk Verifikasi Berita Hoax Christian Khontoro; Justinus Andjarwirawan; Yulia Yulia
Jurnal Infra Vol 9, No 1 (2021)
Publisher : Jurnal Infra

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

Abstract

Hoax or in Indonesian, hoax is fake news or news that has no source. Hoax are a series of information that is misguided, but sold as truth[5]. The problems above are the basis for creating a verification system for this hoax news. The TextRank and Dice Similarity algorithms will be used to help verify the inputed news is a hoax or fact. Where in this study, the TextRank algorithm is used to find the most important keywords in a news which will then be used to become keywords in search engines. Then the Dice Similarity algorithm is used to measure the level of similarity of the news entered with the news obtained from search results on search engines. The hoax verification system that has been done has been tested using several similarity weights to find which similarity weights are the most optimal. The data used were 50 hoax news and 50 fact news. From this test, the optimal similarity weight is 40% with an accuracy of 84%. With details of 50 hoax data, 47 news were declared hoax, 2 news items were declared facts, and 1 news was declared unknown. Of the 50 fact news, 37 news were declared facts, 13 were declared hoax, and no news was declared unknown.
Penerapan Convolutional Neural Network Untuk Klasifikasi Kanker Kulit Melanoma Pada Dataset Gambar Kulit Michael Kurniawan Soegeng; Liliana Liliana; Agustinus Noertjahyana
Jurnal Infra Vol 9, No 1 (2021)
Publisher : Jurnal Infra

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

Abstract

Melanoma skin cancer is one of the most dangerous skin cancers where the ferocity and speed of metastasis has caused a high mortality rate among afflicted when the cancer is not treated. Early detection of the cancer and prevention by removing the affected skin have been shown to decrease the mortality rate on afflicted patient. Thus, development of a method to help automatically diagnose the cancer and classify between cancer and normal mole or birthmark is needed. Previous methods still show limitations in classifying melanoma skin cancer. This study proposes a classification system using convolutional neural network trained on the original ISIC 2020 dataset and hair removed dataset which is then combined using ensemble. The dataset used is first preprocessed using the hair removal algorithm convolutional neural network using EfficientNet B0 – B7 and ResNet-50-v2 will be trained using ISIC 2020 dataset and ISIC 2020 dataset processed with hair removal algorithm.The model is evaluated using test data from ISIC 2020 dataset on area under the receiver operating characteristic curve (ROC AUC). The model trained will then be combined using ensemble where the result of the model will be averaged to give a combined prediction. The result of the test shows that the model trained is capable to classify melanoma and non-melanoma images. It is also shown that by removing hair from the skin image reduces the accuracy of th e model. Using Ensembling on the different models trained into one meta-model also increases the accuracy of the prediction giving a high final accuracy of 93.108%.
Pemanfaatan Simplex Noise Untuk Menghasilkan Map Yang Natural dengan Unity Engine Calvin Vionaldy Tjiandra; Rudy Adipranata; Lily Puspa Dewi
Jurnal Infra Vol 9, No 1 (2021)
Publisher : Jurnal Infra

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

Abstract

In the development of game, terrain is the first thing that theplayer see in the game. Landscape, assets on terrain, and theheight kurvator on the terrain will give an environment that canenhance the experience in playing. But the process of making suchterrain will take a long time because of the terrain setting whichdevelopers do. Moreover, if the developer needs to create manyterrain with a similar parameter. not just that, because thesimilarity from the terrain, the terrain itself will becomerepetitive. To avoid that, it is better to use the Procedural ContentGenerator or PCG to create a base terrain with every project.Simplex noise is one of many PCG method to generate a terrainheight map to be developed by developer. Terrain that had beengenerated can be explore by the player , and to diminish thechance of slump because of the repetitive play, the surface will bedifferent each time the method generate.This paper will be focus on the implementation of simplex noise togenerate a base terran for a 3D game with the usage of C# fromunity. simplex noise will generate 3 biomes which are the forest,savanna, mountain. Perlin noise will be implemented as well forthe purpose of comparison method, perlin noise will generate thesame biome as simplex noise. This program will see the result ofthe surface generated by both methods.This paper will test some terrain models by how the result ofsimplex generated terrain and perlin generated terrain can makea variety for terrain. Terrain will change each time the method isrunning in the program. In the result of the paper, it is found thatperlin have an advantage to generate that have a low to middleheight terrain and simplex have the advantage to generate aterrain that have a middle to high height.

Page 3 of 4 | Total Record : 33

 1   2   3   4 

Filter by Year

2021 2021


Reset
Filter By Issues
All Issue Vol 10, No 2 (2022) Vol 10, No 1 (2022) Vol 9, No 2 (2021) Vol 9, No 1 (2021) Vol 8, No 2 (2020) Vol 8, No 1 (2020) Vol 7, No 2 (2019) Vol 7, No 1 (2019) Vol 6, No 2 (2018) Vol 6, No 2 (2018) Vol 6, No 1 (2018) Vol 5, No 2 (2017) Vol 5, No 1 (2017) Vol 4, No 2 (2016) Vol 4, No 1 (2016) Vol 4, No 1 (2016) Vol 3, No 2 (2015) Vol 3, No 2 (2015) Vol 3, No 1 (2015) Vol 3, No 1 (2015) Vol 2, No 2 (2014) Vol 2, No 2 (2014) Vol 2, No 1 (2014) Vol 2, No 1 (2014) Vol 1, No 2 (2013) Vol 1, No 2 (2013) Vol 1, No 1 (2013) Vol 1, No 1 (2013) More Issue