• Annual Conference on Human and Language Technology
• /
• 2005.10a
• /
• pp.89-96
• /
• 2005

본 연구에서는 영한 기계 번역에서 20단어 이상의 긴 문장을 보다 정확히 분석하기 위하여 문장을 복수개의 의미 있는 절로 분할하고자 한다. 긴 문장은 구문 분석을 시도할 때, 시간적으로 또는 공간적으로 급격히 증가하는 자원을 소모시킨다. 이러한 문제를 해결하기 위하여, 본 연구에서는 긴 문장에서 분할 가능한 지점을 인식하여 이러한 지점을 중심으로 여러 개의 절을 생성한 후, 이 절을 개별적으로 분석하고자 하였다. 문장을 분할하기 위해서 일단 문장 내부에 존재하고 있는 분할이 가능한 지점을 선택하고, 선택된 지점을 중심으로 문맥 정보를 표현하는 입력 벡터를 생성하였다. 그리고 Support Vector Machine (SVM)을 이용하여 이러한 후보 지점의 특성을 학습하여 향후 긴 문장이 입력되었을 때 보다 정확하게 분할점을 찾고자 하였다. 본 논문에서는 SVM의 보다 좋은 학습과 분류를 위하여 내부 커널로써 다항 커널 (polynomial kernel)을 사용하였다. 그리고 실험을 통하여 약 0.97의 f-measure 값을 얻을 수 있었다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2008.06c
• /
• pp.285-288
• /
• 2008

최근의 구문분석 연구는 컴퓨터 성능 향상과 사용 가능한 대량의 구문분석 말뭉치 증가, 견고한 기계학습 기법 개발 등에 힘입어 통계적인 모델 연구가 꾸준히 증가하고 있다. 본 논문에서는 기존에 개발된 다양한 기계학습 기법 중 ME(Maximum Entropy) 모델과 SVM(Support vector machine) 모델을 이용한 한국어 구문분석 방법을 제안한다. 국어정보베이스(KIBS) 구문분석 말뭉치를 가지고 실험한 결과 SVM 모델을 이용한 한국어 구문분석기가 기존의 확률 기반 통계적 한국어 구문분석기의 성능보다도 최대 1.84% 높은 87.46%의 의존관계 결정 정확률을 보였다. 추후 언어지식을 반영한 다양한 자질들을 이용할 경우 성능 향상이 기대된다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2017.11a
• /
• pp.810-812
• /
• 2017

최근 기계학습에 대한 관심이 높아지면서 금융 분야에서는 인공지능을 이용하여 투자 포트폴리오를 제안하는 로보어드바이저(robo-advisor)를 출시하고 있다. 이는 고객에게 저렴한 수수료를 제공하며 높은 접근성, 인건비의 절감 등의 장점으로 이를 도입하여 다양한 상품을 개발하고 있다. 본 연구에서는 머신러닝 알고리즘인 SVM(support vector machine)과 kNN(k-nearest neighbor)을 활용하여 매월 12개월 이전의 KOSPI 지수 데이터를 학습시킨 후 예측하는 투자 시스템을 구현하였다. 실험결과 SVM이 2.90413배의 성적으로 가장 우수했으며 수익률은 Precision(예측정확도)와 비례함을 보였다. 또한 수익곡선은 추세에 따라 유사한 형태를 보인 성과를 도출하였다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2012.11a
• /
• pp.468-471
• /
• 2012

음성을 통한 자동화된 감정 인식은 편리하고 다양한 서비스를 제공할 수 있어 중요한 연구분야라고 할 수 있다. 기계학습의 다양한 알고리즘을 사용하여 감정을 인식하는 연구가 진행되어 왔지만 그 성능은 아직 초보적 단계를 벋어나지 못하고 있는 실정이다. 앞선 연구에서 우리는 비감독 학습 방법으로 감성을 그룹화 하고 이것을 이용하여 다시 감독 학습을 하는 시스템을 소개 하였다. 본 연구에서 우리는 감독 학습 방법에서 사용했던 오류 역전파 알고리즘을 support vector machine(SVM) 으로 변경하고 몇 가지 구조를 변경하여 기능을 개선하였다. 실험을 통하여 성능을 측정하였으며 어느 정도 개선된 결과를 얻을 수 있었다.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.47 no.3
• /
• pp.54-60
• /
• 2010

The local appearance-based method is one of the face recognition methods that divides face image into small areas and extracts features from each area of face image using statistical analysis. It collects classification results of each area and decides identity of a face image using a voting scheme by integrating classification results of each area of a face image. The conventional local appearance-based method divides face images into small pieces and uses all the pieces in recognition process. In this paper, we propose a local appearance-based method that makes use of only the relatively important facial components. The proposed method detects the facial components such as eyes, nose and mouth that differs much from person to person. In doing so, the proposed method detects exact locations of facial components using support vector machines (SVM). Based on the detected facial components, a number of small images that contain the facial parts are constructed. Then it extracts features from each facial component image using principal components analysis (PCA). We compared the performance of the proposed method to those of the conventional methods. The results show that the proposed method outperforms the conventional local appearance-based method while preserving the advantages of the conventional local appearance-based method.

• Journal of the Korea Society of Computer and Information
• /
• v.18 no.11
• /
• pp.31-38
• /
• 2013

Feature normalization as a pre-processing step has been widely used in classification problems to reduce the effect of different scale in each feature dimension and error as a result. Most of the existing methods, however, assume some distribution function on feature distribution. Even worse, existing methods do not use the labels of data points and, as a result, do not guarantee the optimality of the normalization results in classification. In this paper, proposed is a supervised rank normalization which combines rank normalization and a supervised learning technique. The proposed method does not assume any feature distribution like rank normalization and uses class labels of nearest neighbors in classification to reduce error. SVM, in particular, tries to draw a decision boundary in the middle of class overlapping zone, the reduction of data density in that area helps SVM to find a decision boundary reducing generalized error. All the things mentioned above can be verified through experimental results.

• Korean Journal of Remote Sensing
• /
• v.40 no.1
• /
• pp.93-101
• /
• 2024

Accurate field crop classification is essential for various agricultural applications, yet existing methods face challenges due to diverse crop types and complex field conditions. This study aimed to address these issues by combining support vector machine (SVM) models with multi-seasonal unmanned aerial vehicle (UAV) images, texture information extracted from Gray Level Co-occurrence Matrix (GLCM), and RGB spectral data. Twelve high-resolution UAV image captures spanned March-October 2021, while field surveys on three dates provided ground truth data. We focused on data from August (-A), September (-S), and October (-O) images and trained four support vector classifier (SVC) models (SVC-A, SVC-S, SVC-O, SVC-AS) using visual bands and eight GLCM features. Farm maps provided by the Ministry of Agriculture, Food and Rural Affairs proved efficient for open-field crop identification and served as a reference for accuracy comparison. Our analysis showcased the significant impact of hyperparameter tuning (C and gamma) on SVM model performance, requiring careful optimization for each scenario. Importantly, we identified models exhibiting distinct high-accuracy zones, with SVC-O trained on October data achieving the highest overall and individual crop classification accuracy. This success likely stems from its ability to capture distinct texture information from mature crops.Incorporating GLCM features proved highly effective for all models,significantly boosting classification accuracy.Among these features, homogeneity, entropy, and correlation consistently demonstrated the most impactful contribution. However, balancing accuracy with computational efficiency and feature selection remains crucial for practical application. Performance analysis revealed that SVC-O achieved exceptional results in overall and individual crop classification, while soybeans and rice were consistently classified well by all models. Challenges were encountered with cabbage due to its early growth stage and low field cover density. The study demonstrates the potential of utilizing farm maps and GLCM features in conjunction with SVM models for accurate field crop classification. Careful parameter tuning and model selection based on specific scenarios are key for optimizing performance in real-world applications.

• Korean Journal of Remote Sensing
• /
• v.36 no.6_1
• /
• pp.1367-1377
• /
• 2020

The purpose of this study is to assess the seismic vulnerability of buildings in Gyeongju city starting with the earthquake that occurred in the city on September 12, 2016, and produce a seismic vulnerability map. 11 influence factors related to geotechnical, physical, and structural indicators were selected to assess the seismic vulnerability, and these were applied as independent variables. For a dependent variable, location data of the buildings that were actually damaged in the 9.12 Gyeongju Earthquake was used. The assessment model was constructed based on random forest (RF) as a mechanic study method and support vector machine (SVM), and the training and test dataset were randomly selected with a ratio of 70:30. For accuracy verification, the receiver operating characteristic (ROC) curve was used to select an optimum model, and the accuracy of each model appeared to be 1.000 for RF and 0.998 for SVM, respectively. In addition, the prediction accuracy was shown as 0.947 and 0.926 for RF and SVM, respectively. The prediction values of the entire buildings in Gyeongju were derived on the basis of the RF model, and these were graded and used to produce the seismic vulnerability map. As a result of reviewing the distribution of building classes as an administrative unit, Hwangnam, Wolseong, Seondo, and Naenam turned out to be highly vulnerable regions, and Yangbuk, Gangdong, Yangnam, and Gampo turned out to be relatively safer regions.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.47 no.6
• /
• pp.137-145
• /
• 2010

Digital vessel have to accurate and efficient mange the digital data from various sensors in the digital vessel. But, In sensor network, it is difficult to transmit and analyze the entire stream data depending on limited networks, power and processor. Therefore it is suitable to use alternative stream data processing after classifying the continuous stream data. In this paper, We propose efficient processing method that arrange some sensors (temperature, humidity, lighting, voice) and process query based on sliding window for efficient input stream and pre-clustering using multiple Support Vector Machine(SVM) algorithm and manage hash table to summarized information. Processing performance improve as store and search and memory using hash table and usage reduced so maintain hash table in memory. We obtained to efficient result that accuracy rate and processing performance of proposal method using 35,912 data sets.

• Journal of IKEEE
• /
• v.13 no.2
• /
• pp.159-166
• /
• 2009

So far, many researches have been done to retrieve music information using static classification descriptors such as genre and mood. Since static classification descriptors are based on diverse content-based musical features, they are effective in retrieving similar music in terms of such features. However, human emotion or mood transition triggered by music enables more effective and sophisticated query in music retrieval. So far, few works have been done to evaluate the effect of human mood transition by music. Using formal representation of such mood transitions, we can provide personalized service more effectively in the new applications such as music recommendation. In this paper, we first propose our Emotion State Transition Model (ESTM) for describing human mood transition by music and then describe a music classification and recommendation scheme based on the ESTM. In the experiment, diverse content-based features were extracted from music clips, dimensionally reduced by NMF (Non-negative Matrix Factorization, and classified by SVM (Support Vector Machine). In the performance analysis, we achieved average accuracy 67.54% and maximum accuracy 87.78%.