Article Per Year (5 Year)
p-Index From 2020 - 2025
0.835
P-Index
This Author published in this journals
All Journal Jurnal Infra
Alvin Nathaniel Tjondrowiguno
Program Studi Teknik Informatika, Universitas Kristen Petra Surabaya
Computer Science & IT

Published : 5 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 5 Documents
Search

 1 
Pemodelan Lip Reading Bahasa Indonesia Berbasis Visem Menggunakan VGG16 serta Jaro-Winkler Similarity dan Bigram Henry Wicaksono; Liliana Liliana; Alvin Nathaniel Tjondrowiguno
Jurnal Infra Vol 10, No 2 (2022)
Publisher : Universitas Kristen Petra

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

Abstract

Lip reading is a technique used to understand spoken words through visual representation of lip movements. Lip reading has many uses, such as aids for laryngectomy patients and aids for people with hearing disabilities. A research shows that 2.6% of Indonesia’s population has a hearing disability. Thus, lip reading can be a relevant solution in Indonesia. This study aims to model a viseme-based Indonesian lip reading system. The method used in this research is VGG16 which is used as a classifier and Jaro-Winkler similarity and bigram (JW-bigram) which is used as a decoder. The dataset used consists of 25 Indonesian sentences composed of 50 different words and spoken by 12 speakers. The results showed that the lip reading system made using VGG16 and JW-bigram was more effective in terms of accuracy and speed compared to other methods combinations.
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

Abstract

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
Penerapan Long-Short Term Memory dengan Word2Vec Model untuk Mendeteksi Hoax dan Clickbait News pada Berita Online di Indonesia Soni Marko Nathanniel Tannady; Djoni Haryadi Setiabudi; Alvin Nathaniel Tjondrowiguno
Jurnal Infra Vol 10, No 2 (2022)
Publisher : Universitas Kristen Petra

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

Abstract

News has become information that is routinely consumed every day and can be accessed easily as technology develops. However, the easy access of readers to news also opens up space for some people to spread clickbait or hoax news to attract readers' attention for personal gain. To overcome this, one of the efforts that can be realized is with a detection model for clickbait and hoax news with machine learning methods. The method used is Long-Short Term Memory. However, with several additional applications such as adding a dropout layer, implementing a callback function and using k-fold cross validation to overcome the problem of overfitting the model which often occurs in related studies. The built model will be tested in a webpage application where users can detect news labels. On the best testing result, testing accuracy for the clickbait detection model are 72.93% and the hoax detection model are 79.17%.
Pengenalan Rambu Lalu Lintas di Indonesia Secara Realtime Menggunakan YOLOv4-tiny Gregorius Nicholas Goenawan; Alvin Nathaniel Tjondrowiguno; Liliana Liliana
Jurnal Infra Vol 10, No 2 (2022)
Publisher : Universitas Kristen Petra

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

Abstract

Concentration are crucial when driving. Drivers who lose their concentration tend to have a slower reaction time, and a higher possibility of violating traffic signs. Traffic signs violation is considered a criminal act with harsh penalties. In addition, traffic sign violations interferes with comfort and endanger other road users. Therefore, we need a system that is able to detect signs accurately and quickly which can inform driver in advance. A research on traffic signs detection on Swedish and Slovenian traffic signs use Mask R-CNN model which based on convolutional neural networks [18]. These method was capable of achieving a mAP@50 score that exceeds 95%. However, the research did not evaluate on the detection speed of such methods. In this research, YOLOv4-tiny is used to detect Indonesian traffic signs. Dataset used in this research are independently collected, which consist of nine prohibition signs and two command signs. The YOLOv4-tiny method with input size of 416 x 416 is able to achieve mAP@50 score of 88.55% with detection speed of 19.41 FPS. With modification to input size and dataset, YOLOv4-tiny are able to achieve mAP@50 score up to 89.58% and detection speed up to 30.87 FPS. YOLOv4-tiny are also able to detect road signs from distance of around 5 to 15 meters with 80.42 % accuracy. Indonesian traffic sign recognition program made by utilizing the YOLOv4-tiny model achieve average recall of 72.9%.
Penerapan Ensemble Learning Menggunakan Metode Support Vector Machine, Naïve Bayes Classifier, dan Valence Aware Dictionary for Sentiment Reasoning untuk Meningkatkan Akurasi Sentiment Analysis pada Review Aplikasi Google Play Tania Sunyoto; Djoni Haryadi Setiabudi; Alvin Nathaniel Tjondrowiguno
Jurnal Infra Vol 10, No 2 (2022)
Publisher : Universitas Kristen Petra

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

Abstract

In an age where almost everyone owns a smartphone, more and more mobile applications are being developed and distributed to Google Play. To decide which application to download, customers are influenced by ratings and reviews. Reviews provide more information than ratings, but there are so many that they are difficult and take a long time to obtain. The application of sentiment analysis supported by high accuracy in reviews can make it easier for customers to get sentiment information from th e application and help them make decisions to download / use the application or not. This research uses a combination of Naïve Bayes and SVM machine learning models with the VADER lexicon model, then Ensemble Learning is carried out using Majority Voting, Majority Weighted Voting, and Stacking to improve accuracy. The results of this system indicate that by using Ensemble Learning the accuracy result increases but not significantly even decreases from SVM results of 88.88% to 88.87% using Stacking.
Co-Authors Cristine Ferlly Wiyanto Djoni Haryadi Setiabudi Gregorius Nicholas Goenawan Henry Novianus Palit Henry Wicaksono Liliana Liliana Soni Marko Nathanniel Tannady Tania Sunyoto