• Title/Summary/Keyword: 엔트로피 모델

Search Result 154, Processing Time 0.026 seconds

Enriching Core Ontology with Domain Thesaurus (분야 시소러스를 이용한 코아 온톨로지 확장)

  • Huang, Jin-Xia;Shin, Ji-Ae;Choi, Key-Sun
    • Annual Conference on Human and Language Technology
    • /
    • 2007.10a
    • /
    • pp.31-37
    • /
    • 2007
  • 본 논문에서는 분야 시소러스의 개념과 관계를 이용하여 코아 온톨로지를 확장하는 방법을 제안한다. 분야 시소러스의 개념을 코아 온톨로지의 상위 개념으로 분류하고, 시소러스에서의 광의어(Broader Term: BT)-협의어(Narrower Term: NT) 및 광의어-관련어(Related Term: RT)들 사이의 관계는 코아 온톨로지에서 정의한 의미관계로 분류한다. 유사도와 빈도수 기반의 방법으로 개념 분류를 수행하였고, 관계 분류에서는 두 가지 방법을 적용하였는데, (i) 훈련데이터가 부족한 경우를 위하여 규칙기반 방법으로 BT-NT/RT관계를 isa와 기타 관계(non-isa관계)로 분류하고, 패턴기반 방법으로 non-isa관계를 온톨로지를 위한 의미관계로 분류한다. (ii) 훈련데이터를 충분히 가지고 있을 경우, 최대 엔트로피 모델(MEM)을 적용한 분류 방법을 사용하되, kNN방법으로 훈련데이터를 정제하였다. 본 논문에서 제안한 방법으로 시스템을 구축하였고, 실험 결과, 시스템 성능이 사람에 의한 판단 결과와 비교 가능한 수준이었다.

  • PDF

The Recognition of Printed Chinese Characters using Probabilistic VQ Networks and hierarchical Structure (확률적 VQ 네트워크와 계층적 구조를 이용한 인쇄체 한자 인식)

  • Lee, Jang-Hoon;Shon, Young-Woo;Namkung, Jae-Chan
    • The Transactions of the Korea Information Processing Society
    • /
    • v.4 no.7
    • /
    • pp.1881-1892
    • /
    • 1997
  • This paper proposes the method for recognition of printed chinese characters by probabilistic VQ networks and multi-stage recognizer has hierarchical structure. We use modular neural networks, because it is difficult to construct a large-scale neural network. Problems in this procedure are replaced by probabilistic neural network model. And, Confused Characters which have significant ratio of miss-classification are reclassified using the entropy theory. The experimental object consists of 4,619 chinese characters within the KSC5601 code except the same shape but different code. We have 99.33% recognition rate to the training data, and 92.83% to the test data. And, the recognition speed of system is 4-5 characters per second. Then, these results demonstrate the usefulness of our work.

  • PDF

Comparative Analysis of Anomaly Detection Models using AE and Suggestion of Criteria for Determining Outliers

  • Kang, Gun-Ha;Sohn, Jung-Mo;Sim, Gun-Wu
    • Journal of the Korea Society of Computer and Information
    • /
    • v.26 no.8
    • /
    • pp.23-30
    • /
    • 2021
  • In this study, we present a comparative analysis of major autoencoder(AE)-based anomaly detection methods for quality determination in the manufacturing process and a new anomaly discrimination criterion. Due to the characteristics of manufacturing site, anomalous instances are few and their types greatly vary. These properties degrade the performance of an AI-based anomaly detection model using the dataset for both normal and anomalous cases, and incur a lot of time and costs in obtaining additional data for performance improvement. To solve this problem, the studies on AE-based models such as AE and VAE are underway, which perform anomaly detection using only normal data. In this work, based on Convolutional AE, VAE, and Dilated VAE models, statistics on residual images, MSE, and information entropy were selected as outlier discriminant criteria to compare and analyze the performance of each model. In particular, the range value applied to the Convolutional AE model showed the best performance with AUC PRC 0.9570, F1 Score 0.8812 and AUC ROC 0.9548, accuracy 87.60%. This shows a performance improvement of an accuracy about 20%P(Percentage Point) compared to MSE, which was frequently used as a standard for determining outliers, and confirmed that model performance can be improved according to the criteria for determining outliers.

A study on end-to-end speaker diarization system using single-label classification (단일 레이블 분류를 이용한 종단 간 화자 분할 시스템 성능 향상에 관한 연구)

  • Jaehee Jung;Wooil Kim
    • The Journal of the Acoustical Society of Korea
    • /
    • v.42 no.6
    • /
    • pp.536-543
    • /
    • 2023
  • Speaker diarization, which labels for "who spoken when?" in speech with multiple speakers, has been studied on a deep neural network-based end-to-end method for labeling on speech overlap and optimization of speaker diarization models. Most deep neural network-based end-to-end speaker diarization systems perform multi-label classification problem that predicts the labels of all speakers spoken in each frame of speech. However, the performance of the multi-label-based model varies greatly depending on what the threshold is set to. In this paper, it is studied a speaker diarization system using single-label classification so that speaker diarization can be performed without thresholds. The proposed model estimate labels from the output of the model by converting speaker labels into a single label. To consider speaker label permutations in the training, the proposed model is used a combination of Permutation Invariant Training (PIT) loss and cross-entropy loss. In addition, how to add the residual connection structures to model is studied for effective learning of speaker diarization models with deep structures. The experiment used the Librispech database to generate and use simulated noise data for two speakers. When compared with the proposed method and baseline model using the Diarization Error Rate (DER) performance the proposed method can be labeling without threshold, and it has improved performance by about 20.7 %.

Voice Recognition Performance Improvement using the Convergence of Voice signal Feature and Silence Feature Normalization in Cepstrum Feature Distribution (음성 신호 특징과 셉스트럽 특징 분포에서 묵음 특징 정규화를 융합한 음성 인식 성능 향상)

  • Hwang, Jae-Cheon
    • Journal of the Korea Convergence Society
    • /
    • v.8 no.5
    • /
    • pp.13-17
    • /
    • 2017
  • Existing Speech feature extracting method in speech Signal, there are incorrect recognition rates due to incorrect speech which is not clear threshold value. In this article, the modeling method for improving speech recognition performance that combines the feature extraction for speech and silence characteristics normalized to the non-speech. The proposed method is minimized the noise affect, and speech recognition model are convergence of speech signal feature extraction to each speech frame and the silence feature normalization. Also, this method create the original speech signal with energy spectrum similar to entropy, therefore speech noise effects are to receive less of the noise. the performance values are improved in signal to noise ration by the silence feature normalization. We fixed speech and non speech classification standard value in cepstrum For th Performance analysis of the method presented in this paper is showed by comparing the results with CHMM HMM, the recognition rate was improved 2.7%p in the speech dependent and advanced 0.7%p in the speech independent.

A Study on the ISAR Image Reconstruction Algorithm Using Compressive Sensing Theory under Incomplete RCS Data (데이터 손실이 있는 RCS 데이터에서 압축 센싱 이론을 적용한 ISAR 영상 복원 알고리즘 연구)

  • Bae, Ji-Hoon;Kang, Byung-Soo;Kim, Kyung-Tae;Yang, Eun-Jung
    • The Journal of Korean Institute of Electromagnetic Engineering and Science
    • /
    • v.25 no.9
    • /
    • pp.952-958
    • /
    • 2014
  • In this paper, we propose a parametric sparse recovery algorithm(SRA) applied to a radar signal model, based on the compressive sensing(CS), for the ISAR(Inverse Synthetic Aperture Radar) image reconstruction from an incomplete radar-cross-section(RCS) data and for the estimation of rotation rate of a target. As the SRA, the iteratively-reweighted-least-square(IRLS) is combined with the radar signal model including chirp components with unknown chirp rate in the cross-range direction. In addition, the particle swarm optimization(PSO) technique is considered for searching correct parameters related to the rotation rate. Therefore, the parametric SRA based on the IRLS can reconstruct ISAR image and estimate the rotation rate of a target efficiently, although there exists missing data in observed RCS data samples. The performance of the proposed method in terms of image entropy is also compared with that of the traditional interpolation methods for the incomplete RCS data.

Application of Monte Carlo Simulation to Intercalation Electrochemistry I. Thermodynamic Approach to Lithium Intercalation into LiMn2O4 Electrode

  • Kim, Sung-Woo;Pyun, Su-Il
    • Journal of the Korean Electrochemical Society
    • /
    • v.5 no.2
    • /
    • pp.79-85
    • /
    • 2002
  • The present article is concerned with the application of the Monte Carlo simulation to electrochemistry of lithium intercalation from the thermodynamic view point. This article first introduced the fundamental concepts of the ensembles, and Ising and lattice gas models in statistical thermodynamics for the Monte Carlo simulation in brief. Finally the Monte Carlo method based upon the lattice gas model was employed to analyse thermodynamics of the lithium intercalation into the transition metal oxides. Especially we dealt with the thermodynamic properties as the electrode potential curve and the partial molar internal energy and entropy of lithium ion in the case of the $LiMn_2O_4$ electrode, and consequently confirmed the utility of the Monte Carlo method in the field of electrochemistry of the lithium intercalation.

Image-adaptive lossless image compression (영상 적응형 무손실 이미지 압축)

  • OH Hyun-Jong;Won Jong-woo;Jang Euee S.
    • Proceedings of the Korean Society of Broadcast Engineers Conference
    • /
    • 2003.11a
    • /
    • pp.61-64
    • /
    • 2003
  • 무손실 이미지 압축은 (Lossless Image Compression)은 손실이미지 압축(Lossy Image Compression)에 비해, 압축률(compression ratio)은 떨어지지만, 반면 원이미지와 복원이미지가 완전히 일치하므로, 원인이미지의 품질을 그대로 유지학 수 있다. 따라서, 이미지의 품질(Quality)과 압축효율(compression ratio)은 서로 상반된 관계에 있으며, 지금도 좀 더 놀은 압축효과를 얻으려는 여러 무손실 압축 방법이 발표되고 있다. 무손실 이미지 압축은 이미지의 정확성과 정밀성이 요구되는, 의료영양분야에서 가장 널리 쓰이고 있으며, 그밖에, 원본이미지를 기본으로 다른 이미지프로세싱이 필요한 경우, 압축 복원을 반복적으로 수행할 필요가 있을 때, 기타 사진 예술분야, 원격 영상 등 정밀성이 요구되는 분양에서 쓰이고 있다. [7]. 무손실 이미지 압축의 가장 대표적인 CALIC[3]과 JPEG_LS[2]를 들 수 있다. CALIC은 비교적 높은 압축률을 나타내지만, 3-PASS의 과정을 거치는 복잡도가 지적되고 있다. 반면 JPEG-LS는 압축률은 CALIC에 미치지 못하지만 빠른 코딩/디코딩 속도를 보인다. 본 논문에서는 여거 가지의 예측 모드를 두어, 블록단위별로 주변 CONTEXT에 따라, 최상의 예측 모드를 판단하여, 이를 적용, 픽셀의 여러 값을 최소화하였다. 그 후 적응산술 부호기(Adaptive arithmetc coder)를 이용하여, 인코딩을 하였다. 이때 최대 에러값은 64를 넘지 않게 했으며, 또한 8*8블록별로 에러의 최대값을 측정하여 그 값을 $0\~7$까지의 8개의 대표값으로 양자화하는 방법을 통하여 그에 따라 8개의 보호화 심볼 모델중 알맞은 모델에 적용하였다. 이를 통해, 그 소화값의 확률 구간을 대폭 넓힘으로써, 에러 이미지가 가지고 있는 엔트로피에 좀 근접하게 코딩을 할 수 있게 되었다. 이 방법은 실제로 Arithmetic Coder를 이용하는 다른 압축 방법에 그리고 적용할 수 있다. 실험 결과 압축효율은 JPEG-LS보다 약 $5\%$의 압축 성능 개선이 있었으며, CALIC과는 대등한 압축률을 보이며, 부호화/복호화 속도는 CALIC보다 우수한 것으로 나타났다.

  • PDF

Development of Online Machine Learning Model for AHU Supply Air Temperature Prediction using Progressive Sampling and Normalized Mutual Information (점진적 샘플링과 정규 상호정보량을 이용한 온라인 기계학습 공조기 급기온도 예측 모델 개발)

  • Chu, Han-Gyeong;Shin, Han-Sol;Ahn, Ki-Uhn;Ra, Seon-Jung;Park, Cheol Soo
    • Journal of the Architectural Institute of Korea Structure & Construction
    • /
    • v.34 no.6
    • /
    • pp.63-69
    • /
    • 2018
  • The machine learning model can capture the dynamics of building systems with less inputs than the first principle based simulation model. The training data for developing a machine learning model are usually selected in a heuristic manner. In this study, the authors developed a machine learning model which can describe supply air temperature from an AHU in a real office building. For rational reduction of the training data, the progressive sampling method was used. It is found that even though the progressive sampling requires far less training data (n=60) than the offline regular sampling (n=1,799), the MBEs of both models are similar (2.6% vs. 5.4%). In addition, for the update of the machine learning model, the normalized mutual information (NMI) was applied. If the NMI between the simulation output and the measured data is less than 0.2, the model has to be updated. By the use of the NMI, the model can perform better prediction ($5.4%{\rightarrow}1.3%$).

An Optimization on the Psychoacoustic Model for MPEG-2 AAC Encoder (MPEG-2 AAC Encoder의 심리음향 모델 최적화)

  • Park, Jong-Tae;Moon, Kyu-Sung;Rhee, Kang-Hyeon
    • Journal of the Institute of Electronics Engineers of Korea CI
    • /
    • v.38 no.2
    • /
    • pp.33-41
    • /
    • 2001
  • Currently, the compression is one of the most important technology in multimedia society. Audio files arc rapidly propagated throughout internet Among them, the most famous one is MP-3(MPEC-1 Laver3) which can obtain CD tone from 128Kbps, but tone quality is abruptly down below 64Kbps. MPEC-II AAC(Advanccd Audio Coding) is not compatible with MPEG 1, but it has high compression of 1.4 times than MP 3, has max. 7.1 and 96KHz sampling rate. In this paper, we propose an algorithm that decreased the capacity of AAC encoding computation but increased the processing speed by optimizing psychoacoustic model which has enormous amount of computation in MPEG 2 AAC encoder. The optimized psychoacoustic model algorithm was implemented by C++ language. The experiment shows that the psychoacoustic model carries out FFT(Fast Fourier Transform) computation of 3048 point with 44.1 KHz sampling rate for SMR(Signal to Masking Ratio), and each entropy value is inputted to the subband filters for the control of encoder block. The proposed psychoacoustic model is operated with high speed because of optimization of unpredictable value. Also, when we transform unpredictable value into a tonality index, the speed of operation process is increased by a tonality index optimized in high frequency range.

  • PDF