• Journal of Industrial Convergence
• /
• v.20 no.4
• /
• pp.23-30
• /
• 2022

Recently, with the development of artificial intelligence (AI) and deep learning, the importance of conversational artificial intelligence chatbots is being highlighted. In addition, chatbot research is being conducted in various fields. To build a chatbot, it is developed using an open source platform or a commercial platform for ease of development. These chatbot platforms mainly use RNN and application algorithms. The RNN algorithm has the advantages of fast learning speed, ease of monitoring and verification, and good inference performance. In this paper, a method for improving the inference performance of RNNs and applied algorithms was studied. The proposed method used the word group expansion learning technique of key words for each sentence when RNN and applied algorithm were applied. As a result of this study, the RNN, GRU, and LSTM three algorithms with a cyclic structure achieved a minimum of 0.37% and a maximum of 1.25% inference performance improvement. The research results obtained through this study can accelerate the adoption of artificial intelligence chatbots in related industries. In addition, it can contribute to utilizing various RNN application algorithms. In future research, it will be necessary to study the effect of various activation functions on the performance improvement of artificial neural network algorithms.

• The KIPS Transactions:PartA
• /
• v.12A no.6 s.96
• /
• pp.515-522
• /
• 2005

This paper presents a method for effectively modeling inlay/onlay prostheses restoring a tooth that are partially destroyed. An inlay/onlay is composed of internal surface adhering to an abutment, and external surface revealed to the outside sight. Internal surfaces are modeled using Minkowski sum expanding the grinded surface parts of abutments so that the internal surfaces can adhere to the abutments with closer contact. In modeling external surfaces, we exploit 3D mesh deformation techniques: DMFFD(direct manipulation free-form deformation)[19] and MWD(multiple wires deformation)[17] with three kinds of informations: standard teeth models, mesh data obtained by scanning a plaster cast of a patient's tooth, FGP(functionally guided plane) measuring the occlusion of the patients's teeth. The standard teeth models are used for building up the basic shapes of external surfaces, while the plaster fast and FGP data are used for reflecting the unique properties of adjacent md occlusal surfaces of the patients's teeth, which are slightly different to each other but very important for correct functioning. With these informations as input data, the adjacent and occlusal surfaces are automatically generated as mesh data using the techniques of DMFFD and m, respectively. Our method was implemented so that inlay/onlay prostheses fan be designed more accurately by visualizing the generated mesh models with requirements by dentists.

• Journal of KIISE:Databases
• /
• v.35 no.1
• /
• pp.54-66
• /
• 2008

A spatio-temporal join is an expensive operation that is commonly used in spatio-temporal database systems. In order to generate an efficient query plan for the queries involving spatio-temporal join operations, it is crucial to estimate accurate selectivity for the join operations. Given two dataset $S_1,\;S_2$ of discrete data and a timestamp $t_q$, a spatio-temporal join retrieves all pairs of objects that are intersected each other at $t_q$. The selectivity of the join operation equals the number of retrieved pairs divided by the cardinality of the Cartesian product $S_1{\times}S_2$. In this paper, we propose aspatio-temporal histogram to estimate selectivity of spatio-temporal join by extending existing geometric histogram. By using a wide spectrum of both uniform dataset and skewed dataset, it is shown that our proposed method, called Spatio-Temporal Histogram, can accurately estimate the selectivity of spatio-temporal join. Our contributions can be summarized as follows: First, the selectivity estimation of spatio-temporal join for discrete data has been first attempted. Second, we propose an efficient maintenance method that reconstructs histograms using compression of spatial statistical information during the lifespan of discrete data.

• KIPS Transactions on Software and Data Engineering
• /
• v.1 no.2
• /
• pp.109-114
• /
• 2012

Honeycombs are dense structures that small cysts, which generally have about 2~10 mm in diameter, are surrounded by the wall of fibrosis. When honeycomb is found in the patients, the incidence of acute exacerbation is generally very high. Thus, the observation and quantitative measurement of honeycomb are considered as a significant marker for clinical diagnosis. In this point of view, we propose an automatic segmentation method using morphological image processing and assessment of the degree of clustering techniques. Firstly, image noises were removed by the Gaussian filtering and then a morphological dilation method was applied to segment lung regions. Secondly, honeycomb cyst candidates were detected through the 8-neighborhood pixel exploration, and then non-cyst regions were removed using the region growing method and wall pattern testing. Lastly, final honeycomb regions were segmented through the extraction of dense regions which are consisted of two or more cysts using cluster analysis. The proposed method applied to 80 High resolution computed tomography (HRCT) images and achieved a sensitivity of 89.4% and PPV (Positive Predictive Value) of 72.2%.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.11 no.6
• /
• pp.95-102
• /
• 2012

In this paper, we propose a method that can construct efficiently a P2P overlay multicast network composed of many heterogeneous peers in communication bandwidth, processing power and a storage size by selecting a peer in a distributed fashion using an ant-colony theory that is one of the computational intelligence methods. The proposed method considers not only the capacity of a peer but also the number of children peers supported by the peer and the hop distance between a multicast source and the peer when selecting a parent peer of a newly joining node. Thus, an P2P multicast overlay network is constructed efficiently in that the distances between a multicast source and peers are maintained small. In addition, the proposed method works in a distributed fashion in that peers use their local information to find a parent node. Thus, compared to a centralized method where a centralized server maintains and controls the overlay construction process, the proposed method scales well. Through simulations, we show that, by making a few high capacity peers support a lot of low capacity peers, the proposed method can maintain the size of overlay network small even there are a few thousands of peers in the network.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.34 no.2
• /
• pp.207-216
• /
• 2024

As big data was built due to the 4th Industrial Revolution, personalized services increased rapidly. As a result, the amount of personal information collected from online services has increased, and concerns about users' personal information leakage and privacy infringement have increased. Online service providers provide privacy policies to address concerns about privacy infringement of users, but privacy policies are often misused due to the long and complex problem that it is difficult for users to directly identify risk items. Therefore, there is a need for a method that can automatically check whether the privacy policy is safe. However, the safety verification technique of the conventional blacklist and machine learning-based privacy policy has a problem that is difficult to expand or has low accessibility. In this paper, to solve the problem, we propose a safety verification technique for the privacy policy using the GPT-3.5 API, which is a generative artificial intelligence. Classification work can be performed evenin a new environment, and it shows the possibility that the general public without expertise can easily inspect the privacy policy. In the experiment, how accurately the blacklist-based privacy policy and the GPT-based privacy policy classify safe and unsafe sentences and the time spent on classification was measured. According to the experimental results, the proposed technique showed 10.34% higher accuracy on average than the conventional blacklist-based sentence safety verification technique.

• Proceedings of the Korean Information Science Society Conference
• /
• 1998.10a
• /
• pp.424-426
• /
• 1998

공간정보의 효율적 공유를 위해 인터넷 기반 GIS소프트웨어 개발 및 응용과 관련된 연구가 활발히 진행 중에 있다. 여러 인터넷 GIS의 기본적인 요구사항 및 현재까지 개발모델과 문제점을 살펴보고, 표준 인터넷 기술을 기반으로 최근 웹 기술 표준 동향을 포함한, OpenGIS상호 운용성이 지원되는 인터넷 GIS기본 구조를 제시하고자 한다. 표준화될 인터넷 GIS 속도 향상과 TCP/IP상의 보안문제가 해결되어야 하고, OpenGIS에서 구성하고 있는 공간 데이터 공유를 위한 표준 사양을 준수할 뿐 아니라 클라이언트/서버의 부하가 최적화된 구조여야한다. 특히 웹 중심의 각종 인터넷 기술들, 즉 HTTP NG. XML, SSL등의 표준 기술이 함께 적용되어야 한다. 새로운 인프라구조는 GIS D/B에 포함된 확장된 (Enhanced) HTTP/MAP 서버와 클라이언트로 구성된다. MAP클라이언트는 MIME-TYPE 에 따라 GIS데이터를 표시할 수 있는 윈도우 환경으로 변환되며 GIS 데이터셋은 XML을 기반으로 하는 MapML(Mapping Makup Language)를 통해 형식을 정한다. 클라이언트가 MapML 토큐먼트를 통해 정의된 구획의 레이어와 벡터 데이터를 요청하고, Map서버는GIS D/B에서 WKB 혹은 소위 VML 형태로 추출하여 클라이언트로 보내주게 된다. 주어진 구획은 MapML로 정의된 속성들을 통해 각종 부가 정보를 열람할 수 있다. MAP은 HTTP와 같은 형태로 동작하므로 전자인증, 암호화를 통한 GIS정보 보안, 클라이언트와 서버 부하의 효율적인 분배 XML을 통한 다양한 GIS속성표현이 가능하다. 본 구조는 Apache +Amiya + Crass D/B+ MapML 환경에서 구현되고 있다.팔일 전송 기법을 각각 제시하고 실험을 통해 이들의 특성을 비교분석하였다.미에서 uronic acid 함량이 두 배 이상으로 나타났다. 흑미의 uronic acid 함량이 가장 많이 용출된 분획은 sodium hydroxide 부분으로서 hemicellulose구조가 polyuronic acid의 형태인 것으로 사료된다. 추출획분의 구성단당은 여러 곡물연구의 보고와 유사하게 glucose, arabinose, xylose 함량이 대체로 높게 나타났다. 점미가 수가용성분에서 goucose대비 용출함량이 고르게 나타나는 경향을 보였고 흑미는 알칼리가용분에서 glucose가 상당량(0.68%) 포함되고 있음을 보여주었고 arabinose(0.68%), xylose(0.05%)도 다른 종류에 비해서 다량 함유한 것으로 나타났다. 흑미는 총식이섬유 함량이 높고 pectic substances, hemicellulose, uronic acid 함량이 높아서 콜레스테롤 저하 등의 효과가 기대되며 고섬유식품으로서 조리 특성 연구가 필요한 것으로 사료된다.리하였다. 얻어진 소견(所見)은 다음과 같았다. 1. 모년령(母年齡), 임신회수(姙娠回數), 임신기간(姙娠其間), 출산시체중등(出産時體重等)의 제요인(諸要因)은 주산기사망(周産基死亡)에 대(對)하여 통계적(統計的)으로 유의(有意)한 영향을 미치고 있어 $25{\sim}29$세(歲)의 연령군에서, 2번째 임신과 2번째의 출산에서 그리고 만삭의 임신 기간에, 출산시체중(出産時體重) $3.50{\sim}3.99kg$사이의 아이에서 그 주산기사망률(周産基死亡率)이 각각 가장 낮았다. 2. 사산(死産)과 초생아사망(初生兒死亡)을 구분(區分)하여 고려해 볼때 사산(死産)은 모성(母

• Journal of Broadcast Engineering
• /
• v.17 no.2
• /
• pp.374-387
• /
• 2012

In this paper, we implement a new hardware for finding the significant coefficients of a digital hologram and ciphering them using discrete wavelet packet transform (DWPT). Discrete wavelet transform (DWT) and packetization of subbands is used, and the adopted ciphering technique can encrypt the subbands with various robustness based on the level of the wavelet transform and the threshold of subband energy. The hologram encryption consists of two parts; the first is to process DWPT, and the second is to encrypt the coefficients. We propose a lifting based hardware architecture for fast DWPT and block ciphering system with multi-mode for the various types of encryption. The unit cell which calculates the repeated arithmetic with the same structure is proposed and then it is expanded to the lifting kernel hardware. The block ciphering system is configured with three block cipher, AES, SEED and 3DES and encrypt and decrypt data with minimal latency time(minimum 128 clocks, maximum 256 clock) in real time. The information of a digital hologram can be hided by encrypting 0.032% data of all. The implemented hardware used about 200K gates in $0.25{\mu}m$ CMOS library and was stably operated with 165MHz clock frequency in timing simulation.

• The Transactions of the Korea Information Processing Society
• /
• v.7 no.3
• /
• pp.879-890
• /
• 2000

Designing a multilevel database application is a complex process, and the entities and their associated security levels must be represented using an appropriate model unambiguously. It is also important to capture the semantics of a multilevel databse application as accurate and complete as possible. Owing to the focus of the IDEA Methodology for designing the non-secure database applications on the data-intensive systems, the Object Model describes the static structure of the objects in an application and their relationships. That is, the Object Model in the IDEA Methodology is an extended Entity-Relationship model giving a static description of objects. The IDEA Methodology has not been developed the multilevel secure database applications, but by using an existing methodology we could take advantage of the various techniques that have already been developed for that methodology. That is, this way is easier to design the multilevel secure schema than to develop a new model from scratch. This paper adds the security features 새？ Object Model in the IDEA Methodology, and presents the transformation from this model to a multilevel secure object oriented schema. This schema will be the preliminary work which can be the general scheme for the automatic mapping to the various commercial multilevel secure database management system such as Informix-Online/Secure, Trusted ORACLE, and Sybase Secure SQL Server.

• Journal of Navigation and Port Research
• /
• v.46 no.4
• /
• pp.367-374
• /
• 2022

The purpose of this study was to propose a deep learning algorithm that applies to the fault diagnosis of fuel pumps and purifiers of autonomous ships. A deep learning algorithm reflecting the time dependence of the measured signal was configured, and the failure pattern was trained using the vibration signal, measured in the equipment's regular operation and failure state. Considering the sequential time-dependence of deterioration implied in the vibration signal, this study adopts Conv1D with sliding window computation for fault detection. The time dependence was also reflected, by transferring the measured signal from two-dimensional to three-dimensional. Additionally, the optimal values of the hyper-parameters of the Conv1D model were determined, using the grid search technique. Finally, the results show that the proposed data preprocessing method as well as the Conv1D model, can reflect the sequential dependency between the fault and its effect on the measured signal, and appropriately perform anomaly as well as failure detection, of the equipment chosen for application.