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
lontarkomputer@unud.ac.id
Editorial Address
Research institutions and Community Service, University of Udayana, Kampus Bukit Jimbaran Bali
Location
Kota denpasar,
Bali
INDONESIA
Lontar Komputer: Jurnal Ilmiah Teknologi Informasi
Published by Universitas Udayana
ISSN : 20881541     EISSN : 25415832     DOI : 10.24843/LKJITI
Core Subject : Science,
Computer Science & IT
Lontar Komputer [ISSN Print 2088-1541] [ISSN Online 2541-5832] is a journal that focuses on the theory, practice, and methodology of all aspects of technology in the field of computer science and engineering as well as productive and innovative ideas related to new technology and information systems. This journal covers research original of paper that has not been published and has been through the double-blind reviewed journal. Lontar Komputer published three times a year by Research institutions and community service, University of Udayana. Lontar Komputer already indexing in Scientific Journal Impact Factor with impact Value 3.968. Lontar Komputer already indexing in SINTA with score S2 and H-index 5.
Arjuna Subject : -
Articles 226 Documents
 19   20   21   22   23 
Predicting the Number of Passengers in Public Transportation Areas Using the Deep Learning Model LSTM Joko Siswanto; Sri Yulianto Joko Prasetyo; Sutarto Wijono; Evi Maria; Untung Rahardja
Lontar Komputer : Jurnal Ilmiah Teknologi Informasi Vol 15 No 03 (2024): Vol.15, No. 3 December 2024
Publisher : Institute for Research and Community Services, Udayana University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24843/LKJITI.2024.v15.i03.p03

Abstract

Accurate predictions of the number of public transport passengers on buses in each region are crucial for operations. They are required by the planning and management authority for bus public transport. A deep learning-based LSTM prediction model is proposed to predict the number of passengers in 4 bus public transportation areas (central, north, south, and west), evaluated by MSLE, MAPE, and SMAPE with dropout, neuron, and train-test variations. The CSV dataset obtained from Auckland Transport(AT) New Zealand metro patronage report on bus performance(1/01/2019-31/07/2023) is used for evaluation. The best prediction model was obtained from the lowest evaluation value and relatively fast time with a dropout of 0.2, 32 neurons, and train-test 80-20. The prediction model on training and testing data improves with the suitability of tuning for four predictions for the next 12 months with mutual fluctuations. The strong negative correlation is central-south, while the strong positive correlation is north-west. Predictions are less closely interconnected and dependent, namely central-south. With its potential to significantly impact policy-making, this prediction model can increase public transport mobility in each region, leading to a more efficient and accessible public transport system and ultimately enhancing the public's daily lives. This research has practical implications for public transport authorities, as it can guide them in making informed decisions about service planning and resource allocation.
Determining The Ripeness Level Of Crystal Guava Fruit Using Backpropagation Neural Network Shofia Nabila Azzahra; Ahmad Kamsyakawuni; Abduh Riski
Lontar Komputer : Jurnal Ilmiah Teknologi Informasi Vol 15 No 03 (2024): Vol.15, No. 3 December 2024
Publisher : Institute for Research and Community Services, Udayana University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24843/LKJITI.2024.v15.i03.p04

Abstract

The ripeness of crystal guava fruit is currently sorted conventionally by analyzing the colour of the rind visually with the human eye. However, this method has several weaknesses that result in low accuracy and inconsistency. Therefore, automatic determination of ripeness level is necessary to increase accuracy and obtain precise information. This research uses the HSI colour space as an interpretation of fruit image characteristics and uses the Backpropagation algorithm to perform classification. This study utilizes image data of crystal guava fruit, categorizing them into four stages of ripeness: unripe, half-ripe, ripe, and very ripe. There are 140 fruit image data with 35 data for each ripeness category. Each image will be processed with median filter, cropping and segmentation. The HSI value will be taken from the image and processed at the classification stage using the Backpropagation algorithm. In classification using Backpropagation Neural Network, the best network model in this study was achieved in the 3 10 4 network architecture with a binary sigmoid activation function, learning rate = 0.3, and batch size = 64. This model produces a loss value of 0.5364 with an accuracy of 0.9 in testing process.
QnA Chatbot with Mistral 7B and RAG method: Traffic Law Case Study Muhammad Roiful Anam; Agus Subhan Akbar; Heru Saputro
Lontar Komputer : Jurnal Ilmiah Teknologi Informasi Vol 15 No 03 (2024): Vol.15, No. 3 December 2024
Publisher : Institute for Research and Community Services, Udayana University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24843/LKJITI.2024.v15.i03.p06

Abstract

Mistral 7B is a language model designed to achieve high efficiency and performance in handling Natural Language Processing (NLP). This research will evaluate the model's effectiveness in legal data processing using the Retrieval-Augmented Generation (RAG) method, focusing on road traffic and transportation law No 22/2009. The system was built using the LangChain framework, followed by fine-tuning the model and evaluated using BERTScore. Results showed that the fine-tuned Mistral 7B achieved an F1 score of 0.9151, higher than the version without fine-tuning (0.8804) and GPT-4 (0.8364). To improve accuracy, the model utilizes specific keywords that make it easier to find relevant data. Fine-tuning was shown to enhance precision, while the use of key elements in questions helped the model focus more on important information. The results are expected to support the development of artificial intelligence (AI) in Indonesia's legal system and provide practical guidance for applying AI technology in other areas of law.
Comparative Analysis of YOLOv8 and HSV Methods for Traffic Density Measurement Prof. I Gede Pasek Suta Wijaya; Muhamad Nizam Azmi; Ario Yudo Husodo
Lontar Komputer : Jurnal Ilmiah Teknologi Informasi Vol 15 No 02 (2024): Vol. 15, No. 2 August 2024
Publisher : Institute for Research and Community Services, Udayana University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24843/LKJITI.2024.v15.i02.p06

Abstract

Traffic density measurement is a critical component in traffic management and urban planning. This study addresses the challenge of accurately measuring traffic density by comparing the performance of the YOLOv8 segmentation method with the traditional HSV method. At the beginning of the abstract, we clearly present the problem of accurately measuring traffic density. The primary objective is to highlight the strengths and limitations of each method in terms of accuracy and reliability in traffic density estimation. The choice of segmenting the asphalt area rather than vehicle objects is justified by the need to understand how different segmentation approaches affect traffic density measurements. The HSV method involves converting images to the HSV color space, creating masks for specific areas, and measuring traffic density based on the asphalt area. This method, while straightforward, may not accurately capture the dynamic nature of vehicle movement. In contrast, the YOLOv8 segmentation method utilizes a deep learning approach to detect and segment vehicles, providing potentially more precise measurements. Experimental results from three locations demonstrate varying levels of traffic density. The YOLOv8 method results in a graph with a wavy pattern, reflecting the more detailed detection of vehicles. Conversely, the HSV method produces a linear pattern, indicating a more consistent but potentially less detailed measurement. Quantitative analysis shows that Location 2 has a higher traffic density compared to Locations 1 and 3, as indicated by the average number of detected vehicles per frame. This study provides a comprehensive understanding of the differences between HSV and YOLOv8 segmentation methods for traffic density measurement. The findings suggest that while YOLOv8 offers more detailed and dynamic detection, the HSV method provides a simpler yet effective approach for certain applications.
Training VGG16, MobileNetV1 and Simple CNN Models from Scratch for Balinese Inscription Recognition Ida Ayu Putu Febri Imawati; Made Sudarma; I Ketut Gede Darma Putra; I Putu Agung Bayupati; Minho Jo
Lontar Komputer : Jurnal Ilmiah Teknologi Informasi Vol 15 No 03 (2024): Vol.15, No. 3 December 2024
Publisher : Institute for Research and Community Services, Udayana University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24843/LKJITI.2024.v15.i03.p01

Abstract

Many inscriptions in Bali are damaged. Damage to these inscriptions can be caused by natural disasters, overgrown with moss, algae and bacteria. Damage can also be caused by warfare, or deliberately erased. This inscription contains the knowledge and civilization of the ancestors so it is very important to be able to read its contents. Based on these problems, this research conducted training from scratch on 3 CNN models namely VGG16, MobileNetV1 and Simple CNN. The purpose of this research is to choose one recognition model that has the best performance and produces the highest recognition rate to proceed to the inscription restoration stage. The dataset used is Balinese inscription: Isolated Character Recognition of Balinese Script in Palm Leaf Manuscript Images in Challenge-3-ForTrain.zip. The training process of three models with five different training files resulted in the finding that VGG16 has the highest accuracy in the training, testing, and validation process with the least number of epochs. This research contributes to specific datasets, such as the Isolated Character Recognition of Balinese Script using the training process from the beginning of VGG16, involving all stages of the process. It will produce the best model performance compared to the other four training models.
Deep Learning Implementation Using CNN to Classify Bali God Sculpture Pictures Ni Luh Gede Pivin Suwirmayanti; I Wayan Budi Sentana; I Ketut Gede Darma Putra; Made Sudarma; I Made Sukarsa; Komang Budiarta
Lontar Komputer : Jurnal Ilmiah Teknologi Informasi Vol 15 No 02 (2024): Vol. 15, No. 2 August 2024
Publisher : Institute for Research and Community Services, Udayana University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24843/LKJITI.2024.v15.i02.p02

Abstract

Efforts to preserve Balinese culture can be carried out by integrating art and technology as new steps that need to be developed. This research is motivated by the existence of various forms of God statues which have a central role in Balinese culture. The Deep Learning method is proposed because it has unique features that can be extracted automatically. The technique used in Deep Learning is Convolutional Neural Network (CNN). The training process is first performed to perform the classification process, and then the testing process is performed. We compared our CNN model with two other models, AlexNet and ResNet, based on the experimental results. Using a data split of 70%- 30%, our CNN model has the highest accuracy in managing statue image data. Specifically, our model achieves 97.14% accuracy, while Alexnet and Resnet achieve 24.44% and 33.33%, respectively. Apart from contributing to introducing the Balinese God Statue, this research can also be a basis for developing more comprehensive applications in culture and tourism.

Page 23 of 23 | Total Record : 226

 19   20   21   22   23 

Filter by Year

2010 2024


Reset
Filter By Issues
All Issue Vol 15 No 03 (2024): Vol.15, No. 3 December 2024 Vol 15 No 02 (2024): Vol. 15, No. 2 August 2024 Vol 15 No 1 (2024): Vol. 15, No. 1 April 2024 Vol 14 No 3 (2023): Vol. 14, No. 3 December 2023 Vol 14 No 2 (2023): Vol. 14, No. 2 August 2023 Vol 14 No 1 (2023): Vol. 14, No. 1 April 2023 Vol 13 No 3 (2022): Vol. 13, No. 3 December 2022 Vol 13 No 2 (2022): Vol. 13, No. 2 August 2022 Vol 13 No 1 (2022): Vol. 13, No. 1 April 2022 Vol 12 No 3 (2021): Vol. 12, No. 03 December 2021 Vol 12 No 2 (2021): Vol. 12, No. 02 August 2021 Vol 12 No 1 (2021): Vol. 12, No. 01 April 2021 Vol 11 No 3 (2020): Vol. 11, No. 03 December 2020 Vol 11 No 2 (2020): Vol. 11, No. 2 August 2020 Vol 11 No 1 (2020): Vol. 11, No. 1 April 2020 Vol. 10, No. 3 December 2019 Vol. 10, No. 2 August 2019 Vol. 10, No. 1 April 2019 Vol. 9, No. 3 December 2018 Vol. 9, No. 2 August 2018 Vol. 9, No. 1 April 2018 Vol. 8, No. 3 Desember 2017 Vol. 8, No. 2 Agustus 2017 Vol. 8, No. 1 April 2017 Vol. 7, No. 3 Desember 2016 Vol. 7, No. 2 Agustus 2016 Vol. 7, No. 1 April 2016 Vol. 6, No.3 Desember 2015 Vol. 6, No.2 Agustus 2015 Vol. 6, No.1 April 2015 Vol. 5, No. 2 Agustus 2014 Vol. 5, No.1 April 2014 Vol. 4, No.3 Desember 2013 Vol. 4, No. 2 Agustus 2013 Vol. 4, No. 1 April 2013 Vol. 3, No. 2 Desember 2012 Vol. 2, No. 1 Juni 2011 Vol. 1, No. 1 Desember 2010 More Issue