Article Per Year (5 Year)
p-Index From 2021 - 2026
1.784
P-Index
This Author published in this journals
All Journal Jurnal Infra
Liliana Liliana
Program Studi Informatika
Computer Science & IT

Published : 13 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 13 Documents
Search

 1   2 
Klasifikasi dalam Pembuatan Portal Berita Online dengan Menggunakan Metode BERT Jehezkiel Hardwin Tandijaya; Liliana Liliana; Indar Sugiarto
Jurnal Infra Vol 9, No 2 (2021)
Publisher : Jurnal Infra

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

Abstract

Internet helps human by making various information from many online news platform accessible. But nowadays, there are a lot of news that can be accessed in different online news platform and needs to be categorized. The news that can be accessed in some of the sources don’t have high credibility about an event, because the publishers use false and misleading information to push their agendas. So in order to check the credibility of an event, it is needed to also read from other sources and not only from 1 source. However, this is not effective because the reader has to look for another news source with different URL address. In this research scraping will be done to retrieve the news that are available in a news platform. After the scraping process is done, the news will be classified to determine the category of the news. The method that will be used is Bidirectional Encoder Representations from Transformers. From the testing of this research, the news can be retrieved and classified. The testing with a pre-trained model indobenchmark /indobert-base-p1 get a very good result where the accuracy reaches 87.548%.
Kecerdasan Buatan dengan Metode ID3 Finite State Machine dalam Turn-Based Tactics Game Adam Putra Sulaiman; Liliana Liliana; Leo Willyanto Santoso
Jurnal Infra Vol 9, No 2 (2021)
Publisher : Jurnal Infra

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

Abstract

As the world of video game grown and advances there are evenmore those who likes and play them as well. One way to increaseimmersion within video games is by implementing AI (ArtificialIntelligence). AI is an ability within a program or machine thatcan grant said machine/program a power of “thought” so thatthey can do a command without unnecessary human inputs. AndAI had grown constantly significant since the day it wasconceived.However one of AI’s greatest weakness is if implemented toosimplistic within it’s execution, then when someone manage todecode the AI’s pattern, then said player can use this knowledgefor their advantage. And as a consequence, costed the gameimmersion since the game is played not the way it was intended toor become way too easy. To solve said problem, requires aflexible AI that can adapt to the gameplay. Using ID3 decisiontree, AI are expected to be able to using the variable within aTurn-Based Strategy Game to use different options according tothe data it earn. And hopefully with little improvements, thismethod can be used to differing video game genres as well.The test result proven that while ID3 does give some influence forthe AI decision making, during the test run for Turn-Based Gameshown that the Adaptive AI that always chooses the best option ifplayed the same had a WDL overalls (Win-Draw-Lose) of 1:2:0,while Adaptive AI with the ID3 formula and dynamic decisionmaking had a more balanced results of 1:1:1 and random AIalways ended at a disadvantage with 0:0:3. These tests provedthat Adaptive AI provide the most challenge to the player doesn’thave any influence for the early parts of the game. And that theID3 impact were non-existent during the early game butimproving as the game goes on and the data table more varied.
Electrocardiogram Biometrics Recognition Menggunakan Artificial Neural Network William Sim Jayapranata; Rolly Intan; Liliana Liliana
Jurnal Infra Vol 9, No 1 (2021)
Publisher : Jurnal Infra

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

Abstract

Research on biometrics recognition has become popular in the last two decades. Electrocardiogram signal is one among many data that can be used in biometrics recognition purposes. It is unique for each individual, easy to obtain, and hard to forge made electrocardiogram well suited for biometrics recognition. In this research, an identifier will be made using the electrocardiogram signal of each individual.In this research, non-fiducial approach on MIT-BIH Arrhythmia Database from physionet with Artificial Neural Network as classifier was used. Non-sequential classifier offers lower computational complexity compared to sequential classifiers. Non-fiducial approach does not require feature extraction but a method of truncating the signal to each heartbeat is still required. Artificial Neural Network method uses neuron on each layer to classify digitalized electrocardiogram signal data.Experiment result using our method achieved 98.886% accuracy using MIT-BIH Arrhythmia Database. This research demonstrates Artificial Neural Network method capability as non-sequential classifier to identify electrocardiogram with non-fiducial approach.
Co-Authors Adam Putra Sulaiman Agustinus Noertjahyana Andhika Evantia Irawan Daniel Subroto Hans Juwiantho Hendra Winata Indar Sugiarto Jehezkiel Hardwin Tandijaya Kartika Gunadi Kevin Christian Salim Kevin Kevin Leo Willyanto Santoso Lily Eka Sari Michael Kurniawan Soegeng Richard Gozali Rolly Intan Ryan Chandra Kusuma Steven Hans Gunadi Chua William Sim Jayapranata Yohanes Hendra Wijaya Yulia Yulia