• Journal of Practical Engineering Education
• /
• v.14 no.3
• /
• pp.491-498
• /
• 2022

In this paper, as a liberal arts course for non-major students, a case study of recurrent neural network SW practice, which is essential for designing a basic recurrent neural network subject curriculum, was developed. The developed SW practice case focused on understanding the operation principle of the recurrent neural network, and used a spreadsheet to check the entire visualized operation process. The developed recurrent neural network practice case consisted of creating supervised text completion training data, implementing the input layer, hidden layer, state layer (context node), and output layer in sequence, and testing the performance of the recurrent neural network on text data. The recurrent neural network practice case developed in this paper automatically completes words with various numbers of characters. Using the proposed recurrent neural network practice case, it is possible to create an artificial intelligence SW practice case that automatically completes by expanding the maximum number of characters constituting Korean or English words in various ways. Therefore, it can be said that the utilization of this case of basic practice of recurrent neural network is high.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.18 no.5
• /
• pp.747-754
• /
• 2023

Energy-efficient routing protocol is an important task in a wireless sensor network that is used for monitoring and control by wirelessly collecting information obtained from sensor nodes deployed in various environments. Various routing techniques have been studied for this, but it is also necessary to consider WSN environments with specific situations and conditions. In particular, due to topographical characteristics or specific obstacles, a hole where sensor nodes are not deployed may exist in most WSN environments, which may result in inefficient routing or routing failures. In this case, the geographical routing-based hall bypass routing method using GPS functions will form the most efficient path, but sensors with GPS functions have the disadvantage of being expensive and consuming energy. Therefore, we would like to find the boundary node of the hole in a WSN environment with holes through minimal sensor function and propose hole bypass routing through boundary line formation.

• Journal of Information Technology Services
• /
• v.22 no.5
• /
• pp.99-108
• /
• 2023

As the computerization of hospitals becomes more advanced, security issues regarding data generated from various medical devices within hospitals are gradually increasing. For example, because hospital data contains a variety of personal information, attempts to attack it have been continuously made. In order to safely protect data from external attacks, each hospital has formed an internal team to continuously monitor whether the computer network is safely protected. However, there are limits to how humans can monitor attacks that occur on networks within hospitals in real time. Recently, artificial intelligence models have shown excellent performance in detecting outliers. In this paper, an experiment was conducted to verify how well an artificial intelligence model classifies normal and abnormal data in network traffic data generated from medical devices. There are several models used for outlier detection, but among them, Random Forest and Tabnet were used. Tabnet is a deep learning algorithm related to receive and classify structured data. Two algorithms were trained using open traffic network data, and the classification accuracy of the model was measured using test data. As a result, the random forest algorithm showed a classification accuracy of 93%, and Tapnet showed a classification accuracy of 99%. Therefore, it is expected that most outliers that may occur in a hospital network can be detected using an excellent algorithm such as Tabnet.

• International journal of advanced smart convergence
• /
• v.12 no.4
• /
• pp.434-442
• /
• 2023

The relationship between acute kidney injury (AKI) prediction and nephrotoxic drugs, or drugs that adversely affect kidney function, is one that has yet to be explored in the critical care setting. One contributing factor to this gap in research is the limited investigation of drug modalities in the intensive care unit (ICU) context, due to the challenges of processing prescription data into the corresponding drug representations and a lack in the comprehensive understanding of these drug representations. This study addresses this gap by proposing a novel approach that leverages patient prescription data as a modality to improve existing models for AKI prediction. We base our research on Electronic Health Record (EHR) data, extracting the relevant patient prescription information and converting it into the selected drug representation for our research, the extended-connectivity fingerprint (ECFP). Furthermore, we adopt a unique multimodal approach, developing machine learning models and 1D Convolutional Neural Networks (CNN) applied to clinical drug representations, establishing a procedure which has not been used by any previous studies predicting AKI. The findings showcase a notable improvement in AKI prediction through the integration of drug embeddings and other patient cohort features. By using drug features represented as ECFP molecular fingerprints along with common cohort features such as demographics and lab test values, we achieved a considerable improvement in model performance for the AKI prediction task over the baseline model which does not include the drug representations as features, indicating that our distinct approach enhances existing baseline techniques and highlights the relevance of drug data in predicting AKI in the ICU setting.

• Journal of Technologic Dentistry
• /
• v.46 no.2
• /
• pp.36-41
• /
• 2024

Purpose: This study aimed to generate virtual mandibular left first molar teeth using deep convolutional generative adversarial networks (DCGANs) and analyze their matching accuracy with actual tooth morphology to propose a new paradigm for using medical data. Methods: Occlusal surface images of the mandibular left first molar scanned using a dental model scanner were analyzed using DCGANs. Overall, 100 training sets comprising 50 original and 50 background-removed images were created, thus generating 1,000 virtual teeth. These virtual teeth were classified based on the number of cusps and occlusal surface ratio, and subsequently, were analyzed for consistency by expert dental technicians over three rounds of examination. Statistical analysis was conducted using IBM SPSS Statistics ver. 23.0 (IBM), including intraclass correlation coefficient for intrarater reliability, one-way ANOVA, and Tukey's post-hoc analysis. Results: Virtual mandibular left first molars exhibited high consistency in the occlusal surface ratio but varied in other criteria. Moreover, consistency was the highest in the occlusal buccal lingual criteria at 91.9%, whereas discrepancies were observed most in the occusal buccal cusp criteria at 85.5%. Significant differences were observed among all groups (p<0.05). Conclusion: Based on the classification of the virtually generated left mandibular first molar according to several criteria, DCGANs can generate virtual data highly similar to real data. Thus, subsequent research in the dental field, including the development of improved neural network structures, is necessary.

• Journal of Information Processing Systems
• /
• v.20 no.1
• /
• pp.53-66
• /
• 2024

With the continuous advancement of computer technology, deep learning models have emerged as innovative tools in shaping various aspects of architectural design. Recognizing the distinctive perspective of children, which differs significantly from that of adults, this paper contends that conventional standards may not always be the most suitable approach in designing urban structures tailored for children. The primary objective of this study is to leverage neural style networks within the design process, specifically adopting the artistic viewpoint found in children's illustrations. By combining the aesthetic paradigm of urban architecture with inspiration drawn from children's aesthetic preferences, the aim is to unearth more creative and subversive aesthetics that challenge traditional norms. The selected context for exploration is the landmark buildings in Qingdao City, Shandong Province, China. Employing the neural style network, the study uses architectural elements of the chosen buildings as content images while preserving their inherent characteristics. The process involves artistic stylization inspired by classic children's illustrations and images from children's picture books. Acting as a conduit for deep learning technology, the research delves into the prospect of seamlessly integrating architectural design styles with the imaginative world of children's illustrations. The outcomes aim to provide fresh perspectives and effective support for the artistic design of contemporary urban buildings.

• Journal of KIISE:Computer Systems and Theory
• /
• v.34 no.5_6
• /
• pp.202-213
• /
• 2007

A network is an important model widely used in natural and social science as well as engineering. To analyze these networks easily it is necessary that we should layout the features of networks visually. These Graph-Layout researches have been performed recently according to the development of the computer technology. Among them, the Scale-free Network that stands out in these days is widely used in analyzing and understanding the complicated situations in various fields. The Scale-free Network is featured in two points. The first, the number of link(Degree) shows the Power-function distribution. The second, the network has the hub that has multiple links. Consequently, it is important for us to represent the hub visually in Scale-free Network but the existing Graph-layout algorithms only represent clusters for the present. Therefor in this thesis we suggest Graph-layout algorithm that effectively presents the Scale-free network. The Hubity(hub+ity) repulsive force between hubs in suggested algorithm in this thesis is in inverse proportion to the distance, and if the degree of hubs increases in a times the Hubity repulsive force between hubs is ${\alpha}^{\gamma}$ times (${\gamma}$??is a connection line index). Also, if the algorithm has the counter that controls the force in proportion to the total node number and the total link number, The Hubity repulsive force is independent of the scale of a network. The proposed algorithm is compared with Graph-layout algorithm through an experiment. The experimental process is as follows: First of all, make out the hub that exists in the network or not. Check out the connection line index to recognize the existence of hub, and then if the value of connection line index is between 2 and 3, then conclude the Scale-free network that has a hub. And then use the suggested algorithm. In result, We validated that the proposed Graph-layout algorithm showed the Scale-free network more effectively than the existing cluster-centered algorithms[Noack, etc.].

• Journal of Internet Computing and Services
• /
• v.21 no.6
• /
• pp.23-31
• /
• 2020

This study was carried out to generate various images of railroad surfaces with random defects as training data to be better at the detection of defects. Defects on the surface of railroads are caused by various factors such as friction between track binding devices and adjacent tracks and can cause accidents such as broken rails, so railroad maintenance for defects is necessary. Therefore, various researches on defect detection and inspection using image processing or machine learning on railway surface images have been conducted to automate railroad inspection and to reduce railroad maintenance costs. In general, the performance of the image processing analysis method and machine learning technology is affected by the quantity and quality of data. For this reason, some researches require specific devices or vehicles to acquire images of the track surface at regular intervals to obtain a database of various railway surface images. On the contrary, in this study, in order to reduce and improve the operating cost of image acquisition, we constructed the 'Defective Railroad Surface Regeneration Model' by applying the methods presented in the related studies of the Generative Adversarial Network (GAN). Thus, we aimed to detect defects on railroad surface even without a dedicated database. This constructed model is designed to learn to generate the railroad surface combining the different railroad surface textures and the original surface, considering the ground truth of the railroad defects. The generated images of the railroad surface were used as training data in defect detection network, which is based on Fully Convolutional Network (FCN). To validate its performance, we clustered and divided the railroad data into three subsets, one subset as original railroad texture images and the remaining two subsets as another railroad surface texture images. In the first experiment, we used only original texture images for training sets in the defect detection model. And in the second experiment, we trained the generated images that were generated by combining the original images with a few railroad textures of the other images. Each defect detection model was evaluated in terms of 'intersection of union(IoU)' and F1-score measures with ground truths. As a result, the scores increased by about 10~15% when the generated images were used, compared to the case that only the original images were used. This proves that it is possible to detect defects by using the existing data and a few different texture images, even for the railroad surface images in which dedicated training database is not constructed.

• Journal of Internet Computing and Services
• /
• v.15 no.3
• /
• pp.63-69
• /
• 2014

For the safety messages in IEEE 802.11p vehicles network environment(WAVE), strict periodic beacon broadcasting requires status advertisement to assist the driver for safety. WAVE standards apply multiple radios and multiple channels to provide open public road safety services and improve the comfort and efficiency of driving. Although WAVE standards have been proposed multi-channel multi-radio, the standards neither consider the WAVE multi-radio environment nor its effect on the beacon broadcasting. Most of beacon broadcasting is designed to be delivered on only one physical device and one control channel by the WAVE standard. also conflict-free channel assignment of the fewest channels to a given set of radio nodes without causing collision is NP-hard, even with the knowledge of the network topology and all nodes have the same transmission radio. Based on the latest standard IEEE 802.11p and IEEE 1609.4, this paper proposes an interference aware-based channel assignment algorithm with Nash bargaining solution that minimizes interference and increases throughput with wireless mesh network, which is deigned for multi-radio multi-cahnnel structure of WAVE. The proposed algorithm is validated against numerical simulation results and results show that our proposed algorithm is improvements on 8 channels with 3 radios compared to Tabu and random channel allocation algorithm.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.24 no.11
• /
• pp.1055-1063
• /
• 2013

Low noise amplifiers(LNAs) are an integral component of RF receivers and are frequently required to operate at wide frequency bands for various wireless systems. For wideband operation, important performance metrics such as voltage gain, return loss, noise figures and linearity have been carefully investigated and characterized for the proposed LNA. An inductive shunt feedback configuration is successfully employed in the input stage of the proposed LNA which incorporates cascaded networks with a peaking inductor in the buffer stage. Design equations for obtaining low and high input matching frequencies are easily derived, leading to a relatively simple method for circuit implementation. Careful theoretical analysis explains that poles and zeros are characterized and utilized for realizing the wideband response. Linearity is significantly improved because the inductor between gate and drain decreases the third-order harmonics at the output. Fabricated in $0.18{\mu}m$ CMOS process, the chip area of this LNA is $0.202mm^2$, including pads. Measurement results illustrate that input return loss shows less than -7 dB, voltage gain greater than 8 dB, and a little high noise figure around 7~8 dB over 1.5~13 GHz. In addition, good linearity(IIP3) of 2.5 dBm is achieved at 8 GHz and 14 mA of current is consumed from a 1.8 V supply.