• 제목/요약/키워드: Computer science and engineering

Search Result 16,209, Processing Time 0.051 seconds

Classification of Epileptic Seizure Signals Using Wavelet Transform and Hilbert Transform (웨이블릿 변환과 힐버트 변환을 이용한 간질 파형 분류)

  • Lee, Sang-Hong
    • Journal of Digital Convergence
    • /
    • v.14 no.4
    • /
    • pp.277-283
    • /
    • 2016
  • This study proposed new methods to classify normal and epileptic seizure signals from EEG signals using peaks extracted by wavelet transform(WT) and Hilbert transform(HT) based on a neural network with weighted fuzzy membership functions(NEWFM). This study has the following three steps for extracting inputs for NEWFM. In the first step, the WT was used to remove noise from EEG signals. In the second step, the HT was used to extract peaks from the wavelet coefficients. We also selected the peaks bigger than the average of peaks to extract big peaks. In the third step, statistical methods were used to extract 16 features used as inputs for NEWFM from peaks. The proposed methodology shows that accuracy, specificity, and sensitivity are 99.25%, 99.4%, 99% with 16 features, respectively. Improvement in feature selection method in view to enhancing the accuracy is planned as the future work for selecting good features from 16 features.

Relationship among Degree of Time-delay, Input Variables, and Model Predictability in the Development Process of Non-linear Ecological Model in a River Ecosystem (비선형 시계열 하천생태모형 개발과정 중 시간지연단계와 입력변수, 모형 예측성 간 관계평가)

  • Jeong, Kwang-Seuk;Kim, Dong-Kyun;Yoon, Ju-Duk;La, Geung-Hwan;Kim, Hyun-Woo;Joo, Gea-Jae
    • Korean Journal of Ecology and Environment
    • /
    • v.43 no.1
    • /
    • pp.161-167
    • /
    • 2010
  • In this study, we implemented an experimental approach of ecological model development in order to emphasize the importance of input variable selection with respect to time-delayed arrangement between input and output variables. Time-series modeling requires relevant input variable selection for the prediction of a specific output variable (e.g. density of a species). Inadequate variable utility for input often causes increase of model construction time and low efficiency of developed model when applied to real world representation. Therefore, for future prediction, researchers have to decide number of time-delay (e.g. months, weeks or days; t-n) to predict a certain phenomenon at current time t. We prepared a total of 3,900 equation models produced by Time-Series Optimized Genetic Programming (TSOGP) algorithm, for the prediction of monthly averaged density of a potamic phytoplankton species Stephanodiscus hantzschii, considering future prediction from 0- (no future prediction) to 12-months ahead (interval by 1 month; 300 equations per each month-delay). From the investigation of model structure, input variable selectivity was obviously affected by the time-delay arrangement, and the model predictability was related with the type of input variables. From the results, we can conclude that, although Machine Learning (ML) algorithms which have popularly been used in Ecological Informatics (EI) provide high performance in future prediction of ecological entities, the efficiency of models would be lowered unless relevant input variables are selectively used.

Analysis on Characteristics of Representation of Surrealism on Fantasy Online Game's Character (판타지 온라인 게임의 캐릭터에 관한 초현실주의의 표현 특징 분석)

  • Kim, Kyoung-Nam;Lee, Myoun-Jae
    • Journal of Korea Game Society
    • /
    • v.5 no.4
    • /
    • pp.3-12
    • /
    • 2005
  • Many works have been published with using surrealism techniques among the representation characteristics of media space of video and art. It has begun to research on these work. But many of these researches have not cover surrealism techniques in game graphic. So, in this paper, we have analyzed representation characteristics of fantasy online game's character using surrealism techniques in surrealism. In result, there are means to express nonexistent image in real world in surrealism by using such as collage of parts of body, animals, and other object, deformation, and distortion. Also, these methods have been used in game characters of online game based on fantasy. We analyzed on characteristics of fantasy game's character in aesthetic point view. It should be profitable for making creative and favorable game characters and making creative game graphic by increasing the possibility of art techniques usage.

  • PDF

Acoustic model training using self-attention for low-resource speech recognition (저자원 환경의 음성인식을 위한 자기 주의를 활용한 음향 모델 학습)

  • Park, Hosung;Kim, Ji-Hwan
    • The Journal of the Acoustical Society of Korea
    • /
    • v.39 no.5
    • /
    • pp.483-489
    • /
    • 2020
  • This paper proposes acoustic model training using self-attention for low-resource speech recognition. In low-resource speech recognition, it is difficult for acoustic model to distinguish certain phones. For example, plosive /d/ and /t/, plosive /g/ and /k/ and affricate /z/ and /ch/. In acoustic model training, the self-attention generates attention weights from the deep neural network model. In this study, these weights handle the similar pronunciation error for low-resource speech recognition. When the proposed method was applied to Time Delay Neural Network-Output gate Projected Gated Recurrent Unit (TNDD-OPGRU)-based acoustic model, the proposed model showed a 5.98 % word error rate. It shows absolute improvement of 0.74 % compared with TDNN-OPGRU model.

Design of MHEG Engine for Distributed Multimedia/Hypermedia Applications (분산 멀티미디어/하이퍼미디어 응용을 위한 MHEG 엔진 설계)

  • Lee, Se-Hun;Wang, Chang-Jong
    • The Transactions of the Korea Information Processing Society
    • /
    • v.3 no.2
    • /
    • pp.251-266
    • /
    • 1996
  • In this paper, we design MHEG engine that can generate MHEG objects and present It to the users in Multimedia/Hypermedia Applications In the MHEG engine, the transmitted MHEG objects decoded into internal format. For the easy interpretation of MHEG objects, we define internal format as to be matched for each MHEG object. We easily processobjectinformation using the tree data structure because object inheritance and possession can be represented in tree structure. Object inheritance and possession must be represented in the internal format because they used in resolving the reference to external objector data file. The presentation synchronization extracts the synchronization information from MHEG composite objects, representing and controlling heterogencous media associated to spatio- temporal relation. In order to exactly represent the spatio-temporal synchronization included into MHEG composite object, we propose the algorithm that processes synchronization using the message of the synchronization module and the internal objects. MHEG engine proposed in this paper may be basic technology fro multimedia application area using Korea New Net.

  • PDF

Online Partial Evaluation of Actions (액션의 온라인 부분계산)

  • Gang, Hyeon-Gu;Do, Gyeong-Gu
    • Journal of KIISE:Software and Applications
    • /
    • v.26 no.12
    • /
    • pp.1531-1541
    • /
    • 1999
  • 프로그래밍 언어의 의미를 정형적으로 표기하는 기법인 액션의미론을 기반으로 한 컴파일러 생성기는 프로그래밍언어의 액션의미구조가 주어지면 그 언어의 컴파일러를 자동으로 생성한다. 생성된 컴파일러는 먼저 원시 프로그램을 그에 상응하는 액션 프로그램으로 확장한 후, 목적 프로그램으로 컴파일 한다. 여기서 액션 프로그램은 일종의 중간코드로 쓰이므로, 효율적인 목적코드를 생성하기 위해서 중간코드의 성능향상이 필요하다. 본 논문에서는 액션 프로그램을 부분계산을 통해 효율적인 코드로 자동 변환해 주는 온라인 액션 부분계산기를 설계하고 구현한다. 선행 연구된 오프라인 방식에서 전역분석을 하지 않고는 불가능했던 요약캡슐의 몸통, 펼치기의 몸통에 대한 부분계산이 온라인 방법을 사용하면 가능함을 보이고, 명령형 액션의 부분계산도 추가적으로 수행할 수 있도록 확장한다. Abstract Action Semantics is a framework for formally defining the semantics of programming languages. Action semantics-directed compiler generators take an action semantics definition of a programming language and automatically generate a compiler of the language. The generated compiler first expands a source program into an action denotation of the program, and then compiles it to a target code. In these compiler-generation systems, it is important to statically process the expanded action denotation - used as an intermediate code - as much as possible so that the generated compiler can produce better target code. In this paper, we develop an automatic action-transformation method based on online partial evaluation. The previous off-line method was rather weak because it could not partially evaluate actions inside the body of abstraction and unfolding-action without performing separate global analysis. The proposed online method remedies the problem, thus naturally improves the quality of residual actions. Moreover, we also extend the method to partially evaluate imperative actions.

ICT-based Smart Farm Design (ICT 기반의 스마트팜 설계)

  • Shin, Bong-Hi;Jeon, Hye-Kyoung
    • Journal of Convergence for Information Technology
    • /
    • v.10 no.2
    • /
    • pp.15-20
    • /
    • 2020
  • In this paper, we propose an ICT-based smart farm design. At present, the decrease in rural population is naturally inevitable due to the decrease of the total population. The economic burden on each farm grows with increasing labor costs. As a solution to this, the necessity of spreading smart farms using computing resources is emerging. The proposed system utilizes the ICT technology emerging from the Fourth Industrial Revolution. We will use big data analysis to collect a large amount of data and propose a platform for managing collected data and providing efficient services. The proposed platform consists of SOA service layer, middleware layer, resource pool layer and physical resource layer. ICT-based smart farm service can reduce costs and be easy to install and manage because ICT-based smart farm service provides only necessary functions from the user's point of view.

Analysis and Recognition of Depressive Emotion through NLP and Machine Learning (자연어처리와 기계학습을 통한 우울 감정 분석과 인식)

  • Kim, Kyuri;Moon, Jihyun;Oh, Uran
    • The Journal of the Convergence on Culture Technology
    • /
    • v.6 no.2
    • /
    • pp.449-454
    • /
    • 2020
  • This paper proposes a machine learning-based emotion analysis system that detects a user's depression through their SNS posts. We first made a list of keywords related to depression in Korean, then used these to create a training data by crawling Twitter data - 1,297 positive and 1,032 negative tweets in total. Lastly, to identify the best machine learning model for text-based depression detection purposes, we compared RNN, LSTM, and GRU in terms of performance. Our experiment results verified that the GRU model had the accuracy of 92.2%, which is 2~4% higher than other models. We expect that the finding of this paper can be used to prevent depression by analyzing the users' SNS posts.

Logistic Regression Model on the copyright licence diversification through interindividual Digital Contents distribution (개인간 디지털콘텐츠 유통상의 라이선스 다양화에 대한 로지스틱 회귀모형)

  • Suh, Hye-Sun
    • Journal of Digital Convergence
    • /
    • v.14 no.12
    • /
    • pp.27-33
    • /
    • 2016
  • I would like to analyze the customers accommodation availability of the provisional 'smart board,' having specific mode and style, as a circulation platform of digital contents with using a statistic model in order to find a way and means to activate legal circulation of convergence individual products. The smart board means a circulation platform for both users' convenience and copyright protection, by being conveniently able to upload personal convergence digital contents or apply various licence to the uploaded contents according to the purpose of use. Under these premises of the smart board, this paper is going to focus on verifying to find out which factors, such as users' profile attributes, contents using behaviors, awareness of licence and etc, influence on the intention of using the smart board of general users by using a logistic regression model.

A Design of MILENAGE Algorithm-based Mutual Authentication Protocol for The Protection of Initial Identifier in LTE (LTE 환경에서 초기 식별자를 보호하기 위한 MILENAGE 알고리즘 기반의 상호인증)

  • Yoo, Jae-hoe;Kim, Hyung-uk;Jung, Yong-hoon
    • Journal of Venture Innovation
    • /
    • v.2 no.1
    • /
    • pp.13-21
    • /
    • 2019
  • In LTE environment, which is 4th generation mobile communication systems, there is concern about private information exposure by transmitting initial identifier in plain text. This paper suggest mutual authentication protocol, which uses one-time password utilizing challenge-response and AES-based Milenage key generation algorithm, as solution for safe initial identification communication, preventing unique identification information leaking. Milenage key generation algorithm has been used in LTE Security protocol for generating Cipher key, Integrity key, Message Authentication Code. Performance analysis evaluates the suitability of LTE Security protocol and LTE network by comparing LTE Security protocol with proposed protocol about algorithm operation count and Latency.Thus, this paper figures out initial identification communication's weak points of currently used LTE security protocol and complements in accordance with traditional protocol. So, it can be applied for traditional LTE communication on account of providing additional confidentiality to initial identifier.