• Journal of the Korea Safety Management & Science
• /
• v.22 no.1
• /
• pp.9-14
• /
• 2020

Artificial intelligence (AI) has been applied to most industries by enhancing automation and contributing greatly to efficient processes and high-quality production. This research analyzes the applications of AI-based automobile accident prevention systems. It deals with AI-based collision prevention systems that learn information from various sensors attached to cars and AI-based accident detection systems that automatically report accidents to the control center in the event of a collision. Based on the literature review, technological and institutional changes are taking place at the national levels, which recognize the effectiveness of the systems. In addition, start-ups at home and abroad as well as major car manufacturers are in the process of commercializing auto parts equipped with AI-based collision prevention technology.

• Journal of Drive and Control
• /
• v.18 no.1
• /
• pp.24-30
• /
• 2021

There is an increasing interest in condition-based maintenance for the prevention of economic loss due to failure. Moreover, immense research is being carried out in related technologies in the field of construction machinery. In particular, data-based failure diagnosis methods that employ AI (machine & deep learning) algorithms are in the spotlight. In this study, we have focused on the failure diagnosis and mode classification of reduction gear of excavator's travel device by using the AI algorithm. In addition, a remote monitoring system has been developed that can monitor the status of the reduction gear by using the developed diagnosis algorithm. The failure diagnosis algorithm was performed in the process of data acquisition of normal and abnormal under various operating conditions, data processing and analysis by the wavelet transformation, and learning. The developed algorithm was verified based on three-evaluation conditions. Finally, we have built a system that can check the status of the reduction gear of travel devices on the web using the Edge platform, which is embedded with the failure diagnosis algorithm and cloud.

• Journal of Korea Multimedia Society
• /
• v.23 no.12
• /
• pp.1486-1495
• /
• 2020

The analysis of digital microscopy images plays a vital role in computer-aided diagnosis (CAD) and prognosis. The main purpose of this paper is to develop a machine learning technique to predict the histological grades in prostate biopsy. To perform a multiclass classification, an AI-based deep learning algorithm, a multichannel convolutional neural network (MCCNN) was developed by connecting layers with artificial neurons inspired by the human brain system. The histological grades that were used for the analysis are benign, grade 3, grade 4, and grade 5. The proposed approach aims to classify multiple patterns of images extracted from the whole slide image (WSI) of a prostate biopsy based on the Gleason grading system. The Multichannel Convolution Neural Network (MCCNN) model takes three input channels (Red, Green, and Blue) to extract the computational features from each channel and concatenate them for multiclass classification. Stain normalization was carried out for each histological grade to standardize the intensity and contrast level in the image. The proposed model has been trained, validated, and tested with the histopathological images and has achieved an average accuracy of 96.4%, 94.6%, and 95.1%, respectively.

• Journal of the Microelectronics and Packaging Society
• /
• v.29 no.3
• /
• pp.1-12
• /
• 2022

With the rapid development of artificial intelligence technology that gives existing sensors functions similar to human intelligence is drawing attention. Previously, researches were mainly focused on an improvement of fundamental performance indicators as sensors. However, recently, attempts to combine artificial intelligence such as classification and prediction with sensors have been explored. Based on this, intelligent sensor research has been actively reported in almost all kinds of sensing fields such as disease detection, motion detection, and gas sensor. In this paper, we introduce the basic concepts, types, and driving mechanisms of artificial intelligence and review some examples of its use.

• International Journal of Computer Science & Network Security
• /
• v.21 no.7
• /
• pp.350-358
• /
• 2021

In modern years, the performance of the students is analysed with lot of difficulties, which is a very important problem in all the academic institutions. The main idea of this paper is to analyze and evaluate the academic performance of the college students with bipolar disorder by applying data mining classification algorithms using Jupiter Notebook, python tool. This tool has been generally used as a decision-making tool in terms of academic performance of the students. The various classifiers could be logistic regression, random forest classifier gini, random forest classifier entropy, decision tree classifier, K-Neighbours classifier, Ada Boost classifier, Extra Tree Classifier, GaussianNB, BernoulliNB are used. The results of such classification model deals with 13 measures like Accuracy, Precision, Recall, F1 Measure, Sensitivity, Specificity, R Squared, Mean Absolute Error, Mean Squared Error, Root Mean Squared Error, TPR, TNR, FPR and FNR. Therefore, conclusion could be reached that the Decision Tree Classifier is better than that of different algorithms.

• Journal of Korea Society of Digital Industry and Information Management
• /
• v.18 no.3
• /
• pp.9-23
• /
• 2022

This study aims to examine research trends on the Internet of Things and smart city based on papers from the United States, Japan, and Korea. We collected 7113 papers related to the Internet of Things and smart city published from 2016 to 2021 in Elsevier's Scopus. Keyword frequency and centrality analysis were performed based on the abstracts of the collected papers. We found keywords with high frequency of appearance by calculating keyword frequency and identified central research keywords through the centrality analysis by country. As a result of the analysis, research on security, machine learning, and edge computing related to the Internet of Things and smart city were the most central and highly mediating research conducted in each country. As an implication, studies related to deep learning, cybersecurity, and edge computing in Korea have lower degree centrality and betweenness centrality compared to the United States and Japan. To solve the problem it is necessary to combine these studies with various fields. The future research direction is to analyze research trends on the Internet of Things and smart city in various regions such as Europe and China.

• Review of KIISC
• /
• v.32 no.4
• /
• pp.7-17
• /
• 2022

운전자 보조 시스템을 통한 차량의 전자적인 제어를 위하여, 최근 차량에 탑재된 전자 제어 장치 (ECU; Electronic Control Unit)의 개수가 급증하고 있다. ECU는 효율적인 통신을 위해서 차량용 내부 네트워크인 CAN(Controller Area Network)을 이용한다. 하지만 CAN은 기밀성, 무결성, 접근 제어, 인증과 같은 보안 메커니즘이 고려되지 않은 상태로 설계되었기 때문에, 공격자가 네트워크에 쉽게 접근하여 메시지를 도청하거나 주입할 수 있다. 악의적인 메시지 주입은 차량 운전자 및 동승자의 안전에 심각한 피해를 안길 수 있기에, 최근에는 주입된 메시지를 식별하기 위한 침입 탐지 시스템(IDS; Intrusion Detection System)에 대한 연구가 발전해왔다. 특히 최근에는 AI(Artificial Intelligence) 기술을 이용한 IDS가 다수 제안되었다. 그러나 제안되는 기법들은 특정 공격 데이터셋에 한하여 평가되며, 각 기법에 대한 탐지 성능이 공정하게 평가되었는지를 확인하기 위한 평가 프레임워크가 부족한 상황이다. 따라서 본 논문에서는 machine learning/deep learning에 기반하여 제안된 차랑용 IDS 5가지를 선정하고, 기존에 공개된 데이터셋을 이용하여 제안된 기법들에 대한 비교 및 평가를 진행한다. 공격 데이터셋에는 CAN의 대표적인 4가지 공격 유형이 포함되어 있으며, 추가적으로 본 논문에서는 메시지 주기 유형을 활용한 공격 유형을 제안하고 해당 공격에 대한 탐지 성능을 평가한다.

• International Journal of Computer Science & Network Security
• /
• v.23 no.9
• /
• pp.120-128
• /
• 2023

In this work, we have presented ATSA, a hierarchical attention deep learning model for Arabic sentiment analysis. ATSA was proposed by addressing several challenges and limitations that arise when applying the classical models to perform opinion mining in Arabic. Arabic-specific challenges including the morphological complexity and language sparsity were addressed by modeling semantic composition at the Arabic morphological analysis after performing tokenization. ATSA proposed to perform phrase-chunks sentiment embedding to provide a broader set of features that cover syntactic, semantic, and sentiment information. We used phrase structure parser to generate syntactic parse trees that are used as a reference for ATSA. This allowed modeling semantic and sentiment composition following the natural order in which words and phrase-chunks are combined in a sentence. The proposed model was evaluated on three Arabic corpora that correspond to different genres (newswire, online comments, and tweets) and different writing styles (MSA and dialectal Arabic). Experiments showed that each of the proposed contributions in ATSA was able to achieve significant improvement. The combination of all contributions, which makes up for the complete ATSA model, was able to improve the classification accuracy by 3% and 2% on Tweets and Hotel reviews datasets, respectively, compared to the existing models.

• Journal of Korea Society of Digital Industry and Information Management
• /
• v.19 no.1
• /
• pp.91-107
• /
• 2023

With the recent expansion of the commerce ecosystem, a large number of user evaluations have been produced. Accordingly, attempts to create business insights using user evaluation data have been actively made. However, since user evaluation can change after the user experiences the product, it is difficult to say that the analysis based only on reviews immediately after purchase fully reflects the user's evaluation of the product. Moreover, studies conducted so far on user evaluation have overlooked the fact that the length of time a user has used a product can affect the user's product evaluation. Therefore, in this study, we build a model that predicts the direction of change in the user's rating after use from the user's rating and reviews immediately after purchase. In particular, the proposed model reflects the product's period of use in predicting the change direction of the star rating. However, since the posterior information on the duration of product use cannot be used as input in the inference process, we propose a structure that utilizes information about the product's period of use using an auxiliary classifier. As a result of an experiment using 599,889 user evaluation data collected from the shopping platform 'N' company, we confirmed that the proposed model performed better than the existing model in terms of accuracy.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.26 no.1
• /
• pp.44-53
• /
• 2023

Adversarial attacks have received great attentions for their capacity to distract state-of-the-art neural networks by modifying objects in physical domain. Patch-based attack especially have got much attention for its optimization effectiveness and feasible adaptation to any objects to attack neural network-based object detectors. However, despite their strong attack performance, generated patches are strongly perceptible for humans, violating the fundamental assumption of adversarial examples. In this paper, we propose a camouflaged adversarial patch optimization method using military camouflage assessment metrics for naturalistic patch attacks. We also investigate camouflaged attack loss functions, applications of various camouflaged patches on army tank images, and validate the proposed approach with extensive experiments attacking Yolov5 detection model. Our methods produce more natural and realistic looking camouflaged patches while achieving competitive performance.