Digital Library[ Search Result ]
Effective Embedding Techniques for Misbehavior Classification in Vehicular Ad-Hoc Networks
http://doi.org/10.5626/JOK.2024.51.11.970
Vehicular Ad-Hoc Networks (VANET) is a network technology enabling data transmission between vehicles that includes V2X communication, which facilitates the exchange of both external and internal vehicle information based on communication between vehicles, infrastructure, and pedestrians. However, broadcasting data containing faults or attack information within the network can lead to critical issues, making Misbehavior Detection (MBD) systems an essential technology in VANET. While recent studies have increasingly employed machine learning for MBD, the patterns of misbehavior types in VANET often resemble normal behavior, posing challenges for comprehensive and accurate classification. Existing research has suggested a hierarchical classification system to categorize misbehaviors based on different types of attacks and faults. This study proposed an embedding representation method for constructing a hierarchical classification system to improve the accuracy of misbehavior classification models. By extracting embedding vectors for multivariate time-series data through a pre-trained LSTM model, this study compressed core data related to misbehavior types and employed hierarchical clustering to group various attack types into broader categories.
A Study on Development Method for BERT-based False Alarm Classification Model in Weapon System Software Static Test
Hyoju Nam, Insub Lee, Namhoon Jung, Seongyun Jeong, Kyutae Cho, Sungkyu Noh
http://doi.org/10.5626/JOK.2024.51.7.620
Recently, as the size and complexity of software in weapon systems have increased, securing the reliability and stability is required. To achieve this, developers perform static and dynamic reliability testing during development. However, a lot of false alarms occur in static testing progress that cause wasting resources such as time and cost for reconsider them. Recent studies have tried to solve this problem by using models such as SVM and LSTM. However, they have a critical limitation in that these models do not reflect correlation between defect code line and other lines since they use Word2Vec-based code embedding or only code information. The BERT-based model learns the front-to-back relationship between sentences through the application of a bidirectional transformer. Therefore, it can be used to classify false alarms by analyzing the relationship between code. In this paper, we proposed a method for developing a false alarm classification model using a BERT-based model to efficiently analyze static test results. We demonstrated the ability of the proposed method to generate a dataset in a development environment and showed the superiority of our model.
Unified Prediction of Pedestrian Intention to Jaywalk Based on Parallel Deep Learning Scheme
http://doi.org/10.5626/JOK.2024.51.6.545
Urbanization has led to diversification in traffic accidents and parking issues, with pedestrian accidents at crosswalks accounting for over 30% of traffic fatalities. Particularly concerning are situations where pedestrians are not anticipated by drivers during red signal conditions, as the potential for severe injuries is high. To address this issue, we propose a deep learning-based integrated pedestrian crossing intent prediction system. The system uses the YOLOv5 object detection model to identify pedestrian actions that indicate crossing intent. At the same time, it utilzes the MMPOSE joint prediction model to classify the pedestrian's perspective. By analyzing pedestrian actions, perspectives, and the distance between the pedestrian and the crosswalk, the system predicts crossing intent in various scenarios. Future research based on this study is expected to contribute to diverse application studies aimed at enhancing traffic safety in autonomous driving.
Application of OOD Detection Using MSP in EEG-Based Emotion Classification
HyoSeon Choi, Dahoon Choi, Byung Hyung Kim
http://doi.org/10.5626/JOK.2024.51.5.438
Several deep learning approaches have recently improved the performance of emotion classification tasks. However, these successful applications cannot be directly applied to learning EEG signals because of their nonlinear and complex data structure. This limitation leads to inter- and intra-subject variability problems for understanding complex emotion dynamics. To address this limitation, we focus on studying the variability rather than extracting features from high-dimensional neural activities. In the context of deep learning, we propose a framework to detect and remove abnormal pairs of EEG data and labels for enhancing model performance by utilizing the Maximum Softmax Probability approach. Experimental results on public datasets demonstrated the superiority of our method with a maximum improvement of 4% in accuracy.
Proposal of An Intent Classification Method Using Text Augmentation Techniques and Transfer Learning
Huiwon Lee, Sungho Park, Chaewon Lee, Seunghyun Lee, Kangbae Lee
http://doi.org/10.5626/JOK.2024.51.2.141
Intent classification is the first step of task-directed chatbots and is an important phase in performance improvement. However, task-oriented chatbots are limited by a lack of data for specific domains. The purpose of this study is to solve the problem of data limitation by utilizing text augmentation techniques and transfer learning. Previously, studies using transfer learning and text augmentation techniques existed, but it was difficult to find studies applicable to various domains. This study proposes a text augmentation technique and transfer learning method applicable to various domains. For the experiment, less than 10,000, 20,000, and 30,000 data were constructed according to the ratio of actual utterance intentions in 8 domains. As a result of the experiment, although differences existed depending on the domain, it was confirmed that the method proposed in this study was excellent for all 8 domains. It was confirmed that the accuracy for the 8 domains improved by 10%, 3.4%, and 1.9%, respectively on average with the decreasing size of the training data, and the F1-Score improved by 30%, 12%, and 7.5%, respectively on average.
Hierarchical Representation and Label Embedding for Semantic Classification of Domestic Research Paper
Heejin Kook, Yeonghwa Kim, Sehui Yoon, Byungha Kang, Youhyun Shin
http://doi.org/10.5626/JOK.2024.51.1.41
The sentence"s meaning in the paper is that it has a hierarchical structure, and there is data imbalance among subcategories. In addition, the meaning of the sentence in the paper is closely related to its position within the paper. Existing flat classification methods mainly consider only subcategories, leading to a decrease in classification accuracy due to data imbalance. In response to this, this study proposes hierarchical representation and label embedding methods to perform hierarchical semantic classification of sentences effectively. In addition, the section names of the paper are actively utilized to represent the positional information of the paper sentences. Through experiments, it is demonstrated that the proposed method, which explicitly considers hierarchical and positional information in the KISTI domestic paper sentence semantic tagging dataset, achieves excellent performance in terms of F1 score.
Exploring Neural Network Models for Road Classification in Personal Mobility Assistants: A Comparative Study on Accuracy and Computational Efficiency
Gwanghee Lee, Sangjun Moon, Kyoungson Jhang
http://doi.org/10.5626/JOK.2023.50.12.1083
With the increasing use of personal mobility devices, the frequency of traffic accidents has also risen, with most accidents resulting from collisions with cars or pedestrians. Notably, the compliance rate of the traffic rules on the roads is low. Auxiliary systems that recognize and provide information about roads could help reduce the number of accidents. Since road images have distinct material characteristics, models studied in the field of image classification are suitable for application. In this study, we compared the performance of various road image classification models with parameter counts ranging from 2 million to 30 million, enabling the selection of the appropriate model based on the situation. The majority of the models achieved an accuracy of over 95%, with most models surpassing 99% in the top-2 accuracy. Of the models, MobileNet v2 had the fewest parameters while still exhibiting excellent performance and EfficientNet had stable accuracy across all classes, surpassing 90% accuracy.
The Dataset and a Pretrained Language Model for Sentence Classification in Korean Science and Technology Abstracts
Hongbi Ahn, Soyoung Park, Yuchul Jung
http://doi.org/10.5626/JOK.2023.50.6.468
Classifying each sentence according to its role or function is a critical task, particularly in science and technology papers where abstracts contain various types of research-related content. Proper content curation and appropriate meaning tags are necessary but challenging due to the complexity and diversity of the work. For instance, in biomedical-related abstract data (such as PubMed) in foreign languages, the sentences in the abstract typically follow a consistent semantic sequence, such as background-purpose-method-result-conclusion. However, in Korean paper abstracts, the sentences are described in different orders depending on the author. To address this, we have constructed a dataset (PubKorSci-1k) that tags each sentence according to its role in the abstracts of the science and technology domains described in Korean. Additionally, we propose a learning technique for sentence classification based on this dataset.
A Deep Learning based Speech Quality Enhancement Scheme Using Environmental Sound Classification and Location Information
http://doi.org/10.5626/JOK.2023.50.4.344
In the field of speech processing, deep learning has made great advances by improving the precision of speech recognition. One of advances, voice improvement, is a technique that can improve voice recognition by separating voice and noise from input mixed with speaking voice and noise. This is used in AI-speakers and smartphones to facilitate human-to-human communication and enable clean voice data collection for robots and text-to-speech. However, conventional speech enhancement techniques that use only a single model are not effective in eliminating noise that occurs specifically in each environment. To effectively eliminate environmental specific noise, this paper proposes a deep learning model that combines acoustic scene classification techniques with location information utilization techniques to enable optimal environmental-specific speech enhancements. As a result of the experiment, it is confirmed that this technique shows high voice quality improvement with low computational cost in various environments compared to the existing technique.
Attack Success Rate Analysis of Adversarial Patch in Physical Environment
Hyeon-Jae Jeong, Jubin Lee, Yu Seung Ma, Seung-Ik Lee
http://doi.org/10.5626/JOK.2023.50.2.185
Adversarial patches are widely known as representative adversarial example attacks in physical environment. However, most studies on adversarial patches have demonstrated robust attack success rates based on digital environment rather than physical environment. This study investigated the robustness of adversarial patches in physical environment. To this end, 5 types of generation conditions and 3 types of attachment conditions were derived. The attack success rates of digital patches in physical environment were reviewed according to the changes in conditions. As a basic condition, location, angle, and size variables were targeted as presented in the original adversarial patch paper. Additionally, learning epoch, intent class, and neural network under simulated attack were newly considered and tested as digital patch generation conditions. As a result, the condition which greatly influenced the attack success rates of digital patches was the size. As a learning condition for digital patch generation, digital patches showed sufficient attack success rates with only one to two small learning epochs and simple intent class images. In conclusion, the attack success rate of digital patches in physical environment was not robust unlike in the digital environment.
Search
Journal of KIISE
- ISSN : 2383-630X(Print)
- ISSN : 2383-6296(Electronic)
- KCI Accredited Journal
Editorial Office
- Tel. +82-2-588-9240
- Fax. +82-2-521-1352
- E-mail. chwoo@kiise.or.kr