• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.23 no.4
• /
• pp.115-119
• /
• 2023

The natural language processing models used in current artificial intelligence are huge, causing various difficulties in processing and analyzing data in real time. In order to solve these difficulties, we proposed a method to improve the efficiency of processing by using less memory and checked the performance of the proposed model. The technique applied in this paper to evaluate the performance of the proposed model is to divide the large corpus by adjusting the number of attention heads and embedding size of the BERT[1] model to be small, and the results are calculated by averaging the output values of each forward. In this process, a random offset was assigned to the sentences at every epoch to provide diversity in the input data. The model was then fine-tuned for classification. We found that the split processing model was about 12% less accurate than the unsplit model, but the number of parameters in the model was reduced by 56%.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.05a
• /
• pp.535-538
• /
• 2022

For Plug & Play of IoT devices, we develop a module that recognizes the type of USB, which is a typical wired interface of IoT devices, through image recognition. In order to drive an IoT device, a driver for communication and device hardware is required. The wired interface for connecting to the IoT device is recognized by using the image obtained through the camera of smartphone shooting to recognize the corresponding communication interface. For USB, which is a most popular wired interface, types of USB are classified through artificial neural network-based machine learning. In order to secure sufficient data set of artificial neural networks, USB images are collected through the Internet, and additional image data sets are secured through image processing. In addition to the convolution neural networks, recognizers are implemented with various deep artificial neural networks, and their performance is compared and evaluated.

• Smart Media Journal
• /
• v.11 no.1
• /
• pp.17-30
• /
• 2022

Like other computer vision research trends, research using satellite images was able to achieve rapid growth with the development of GPU-based computer computing capabilities and deep learning methodologies related to image processing. As a result, satellite images are being used in various fields, and the number of studies on how to use satellite images is increasing. Therefore, in this paper, we will introduce the field of research and utilization of satellite images and datasets that can be used for research using satellite images. First, studies using satellite images were collected and classified according to the research method. It was largely classified into a Regression-based Approach and a Classification-based Approach, and the papers used by other methods were summarized. Next, the datasets used in studies using satellite images were summarized. This study proposes information on datasets and methods of use in research. In addition, it introduces how to organize and utilize domestic satellite image datasets that were recently opened by AI hub. In addition, I would like to briefly examine the limitations of satellite image-related research and future trends.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.22 no.6
• /
• pp.51-56
• /
• 2022

With the development of industrialization, the amount of food and waste is rapidly increasing. Accordingly, the government is aware of the seriousness and is making efforts in various ways to reduce it. As a part of that, the volume-based food system was introduced, and although there were several trials and errors at the beginning of the introduction, it shows a reduction effect of 20 to 30%. These results suggest that the volume-based food system is being established. However, the waste is caused by foreign substances in the process of recycling resources by collecting them from the 1st collection to the 2nd collection process. Therefore, in this study, to solve these problems fundamentally, artificial intelligence is applied to classify foreign substances and improve them. Due to the nature of food waste, there is a limit to obtaining many images, so we compare several models based on CNNs and classify them as abnormal data, that is, CNN-based models are trained on various types of foreign substances, and then models with high accuracy are selected. We intend to prepare improvement measures for maintenance, such as manpower input to protect equipment and classify foreign substances by applying it.

• Asia-pacific Journal of Multimedia Services Convergent with Art, Humanities, and Sociology
• /
• v.7 no.10
• /
• pp.935-943
• /
• 2017

CNN (Convolution neural network), which is used for image classification and speech recognition among neural networks learning based on positive data, has been continuously developed to have a high performance structure to date. There are many difficulties to utilize in an embedded system with limited resources. Therefore, we use GPU (General-Purpose Computing on Graphics Processing Units), which is used for general-purpose operation of GPU to solve the problem because we use pre-learned weights but there are still limitations. Since CNN performs simple and iterative operations, the computation speed varies greatly depending on the thread allocation and utilization method in the Single Instruction Multiple Thread (SIMT) based GPGPU. To solve this problem, there is a thread that needs to be relaxed when performing Convolution and Pooling operations with threads. The remaining threads have increased the operation speed by using the method used in the following feature maps and kernel calculations.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.6
• /
• pp.1706-1727
• /
• 2023

Under Water Sensor Networks (UWSN) has gained attraction among various communities for its potential applications like acoustic monitoring, 3D mapping, tsunami detection, oil spill monitoring, and target tracking. Unlike terrestrial sensor networks, it performs an acoustic mode of communication to carry out collaborative tasks. Typically, surface sink nodes are deployed for aggregating acoustic phenomena collected from the underwater sensors through the multi-hop path. In this context, UWSN is constrained by factors such as lower bandwidth, high propagation delay, and limited battery power. Also, the vulnerabilities to compromise the aquatic environment are in growing numbers. The paper proposes an Energy-Efficient standalone Intrusion Detection System (EEIDS) to entail the acoustic environment against malicious attacks and improve the network lifetime. In EEIDS, attributes such as node ID, residual energy, and depth value are verified for forwarding the data packets in a secured path and stabilizing the nodes' energy levels. Initially, for each node, three agents are modeled to perform the assigned responsibilities. For instance, ID agent verifies the node's authentication of the node, EN agent checks for the residual energy of the node, and D agent substantiates the depth value of each node. Next, the classification of normal and malevolent nodes is performed by determining the score for each node. Furthermore, the proposed system utilizes the sheep-flock heredity algorithm to validate the input attributes using the optimized probability values stored in the training dataset. This assists in finding out the best-fit motes in the UWSN. Significantly, the proposed system detects and isolates the malicious nodes with tampered credentials and nodes with lower residual energy in minimal time. The parameters such as the time taken for malicious node detection, network lifetime, energy consumption, and delivery ratio are investigated using simulation tools. Comparison results show that the proposed EEIDS outperforms the existing acoustic security systems.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.5
• /
• pp.1433-1449
• /
• 2023

In recent years, the rapid development of social networks has led to a rapid increase in the amount of information available on the Internet, which contains a large amount of sensitive information related to pornography, politics, and terrorism. In the aspect of sensitive image detection, the existing machine learning algorithms are confronted with problems such as large model size, long training time, and slow detection speed when auditing and supervising. In order to detect sensitive images more accurately and quickly, this paper proposes a multiclassification sensitive image detection method based on lightweight Convolutional Neural Network. On the basis of the EfficientNet model, this method combines the Ghost Module idea of the GhostNet model and adds the SE channel attention mechanism in the Ghost Module for feature extraction training. The experimental results on the sensitive image data set constructed in this paper show that the accuracy of the proposed method in sensitive information detection is 94.46% higher than that of the similar methods. Then, the model is pruned through an ablation experiment, and the activation function is replaced by Hard-Swish, which reduces the parameters of the original model by 54.67%. Under the condition of ensuring accuracy, the detection time of a single image is reduced from 8.88ms to 6.37ms. The results of the experiment demonstrate that the method put forward has successfully enhanced the precision of identifying multi-class sensitive images, significantly decreased the number of parameters in the model, and achieved higher accuracy than comparable algorithms while using a more lightweight model design.

• 한국정보교육학회:학술대회논문집
• /
• 2021.08a
• /
• pp.167-173
• /
• 2021

The purpose of this study is to discuss the types of algorithms and data categories in AI education for elementary school students. The study surveyed 11 pre-elementary teachers after providing education and practice on various data, artificial intelligence algorithm, and AI education platform for 15 weeks. The categories of data and algorithms considering the elementary school level, and educational tools were presented, and their suitability was analyzed. Through the questionnaire, it was concluded that it is most suitable for the teacher to select and preprocess data in advance according to the purpose of the class, and the classification and prediction algorithms are suitable for elementary AI education. In addition, it was confirmed that Entry is most suitable as an AI educational tool, and materials that explain mathematical knowledge are needed to educate the concept of learning of AI. This study is meaningful in that it specifically presents the categories of algorithms and data with in AI education for elementary school students, and analyzes the need for related mathematics education and appropriate AI educational tools.

• Physical Therapy Rehabilitation Science
• /
• v.12 no.2
• /
• pp.80-91
• /
• 2023

Objective: The purpose of this study is to identify factors that affect the incidence of hypertension using logistic regression and decision tree analysis, and to build and compare predictive models. Design: Secondary data analysis study Methods: We analyzed 9,859 subjects from the Korean health panel annual 2019 data provided by the Korea Institute for Health and Social Affairs and National Health Insurance Service. Frequency analysis, chi-square test, binary logistic regression, and decision tree analysis were performed on the data. Results: In logistic regression analysis, those who were 60 years of age or older (Odds ratio, OR=68.801, p<0.001), those who were divorced/widowhood/separated (OR=1.377, p<0.001), those who graduated from middle school or younger (OR=1, reference), those who did not walk at all (OR=1, reference), those who were obese (OR=5.109, p<0.001), and those who had poor subjective health status (OR=2.163, p<0.001) were more likely to develop hypertension. In the decision tree, those over 60 years of age, overweight or obese, and those who graduated from middle school or younger had the highest probability of developing hypertension at 83.3%. Logistic regression analysis showed a specificity of 85.3% and sensitivity of 47.9%; while decision tree analysis showed a specificity of 81.9% and sensitivity of 52.9%. In classification accuracy, logistic regression and decision tree analysis showed 73.6% and 72.6% prediction, respectively. Conclusions: Both logistic regression and decision tree analysis were adequate to explain the predictive model. It is thought that both analysis methods can be used as useful data for constructing a predictive model for hypertension.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2022.05a
• /
• pp.360-360
• /
• 2022

토양수분과 유출 간 관계를 정량화하는 것은 수문 기작 및 유출 발생 과정의 이해를 위한 중요한 정보를 제공한다. 특히, 유출과정의 특성화는 수문 사상에 따른 불포화대 내 토양수 및 토사 손실 제어와 산사태 및 비점오염원 발생 예측을 위해 필수적이다. 유출과정과 관련된 비선형성과 복잡성을 확인하기 위해 토양수분과 유출 사이의 이력 거동이 조사되었다. 특히, 수문 과정 내 이력 현상 구체화를 위해 정성적인 시각적 분류 및 정량적 평가를 위한 이력 지수들이 개발되었다. 정성적인 시각적 분류는 시간에 따라 시계 및 반시계방향으로 다중 루프 형상을 나누는 방식으로 진행되었고, 정량적 평가의 경우 이력 고리(Hysteretic loop) 내 상승 고리(Rising limb)와 하강 고리(Falling limb)의 차이를 기준으로 한 지수로 이력 현상을 특성화하였다. 이전에 제안된 방법론들은 연구자의 판단이 들어가기 때문에 보편적이지 않고 이력 현상을 개발된 지수에 맞춤에 따라 자료 손실이 나타나는 한계가 존재한다. 자료의 손실 없이 불포화대 내 발생 가능한 대표 이력 현상을 자동으로 추출하기 위해 적합한 비지도 학습기반 기계학습 방법론의 제안이 필요하다. 우리 연구에서는 국내 산지 사면에서 강우 사상 동안 다중 깊이(10, 30, 60cm)로 56개의 토양수분 측정지점에서 확보된 토양수분 시계열 자료와 산지 사면 내 위어를 통해 확보된 유출 시계열 자료를 사용하였다. 먼저, 기존에 분류 방법을 기반으로 계절 및 공간특성에 따라 지배적으로 발생하는 토양수분-유출 간 이력 현상을 특성화하였다. 다음으로, 토양수분-유출 간 이력 패턴을 자료 손실 없이 형상화하여 자동으로 데이터베이스화하는 알고리즘을 개발하였다. 마지막으로, 비지도 학습방법을 이용하여 데이터베이스화된 실제 발현 이력 현상 내 확률분포를 최대한 가깝게 추정하는 은닉층을 반복적인 재구성 학습을 통해 구현함으로써 대표 이력 현상 패턴을 추출하였다.