• Journal of information and communication convergence engineering
• /
• 제5권3호
• /
• pp.233-238
• /
• 2007

An approach to identify Chinese event types is proposed in this paper which combines a good feature selection policy and a Maximum Entropy (ME) model. The approach not only effectively alleviates the problem that classifier performs poorly on the small and difficult types, but improve overall performance. Experiments on the ACE2005 corpus show that performance is satisfying with the 83.5% macro - average F measure. The main characters and ideas of the approach are: (1) Optimal feature set is built for each type according to local feature selection, which fully ensures the performance of each type. (2) Positive and negative features are explicitly discriminated and combined by using one - sided metrics, which makes use of both features' advantages. (3) Wrapper methods are used to search new features and evaluate the various feature subsets to obtain the optimal feature subset.

• 한국전자통신학회논문지
• /
• 제17권1호
• /
• pp.153-160
• /
• 2022

그래프신호가 각각의 가중치를 갖고 발생하는 경우 그래프상의 최적의 샘플링 노드집합을 선택하는 탐욕알고리즘에 대해 연구한다. 이를 위해 가중치를 반영한 복원오차를 비용함수로 사용하고 여기에 QR 분해를 적용하여 단순한 형태로 전개한다. 이렇게 도출된 가중치 복원오차를 최소화하기 위해 다양한 수학적 증명을 통해 반복적으로 노드를 선택할 수 있는 수학적 결과식을 유도한다. 이러한 결과식에 기반하여, 노드를 선택하는 샘플링 집합 선택알고리즘을 제안한다. 성능평가를 위해 다양한 그래프에서 발생하는 가중치를 갖는 그래프신호에 적용하여 기존 샘플링 선택 기술대비, 복잡도를 유지하면서 가중치 신호의 복원성능이 우수함을 보인다.

• International Journal of Computer Science & Network Security
• /
• 제23권7호
• /
• pp.131-140
• /
• 2023

Network security situational awareness systems helps in better managing the security concerns of a network, by monitoring for any anomalies in the network connections and recommending remedial actions upon detecting an attack. An Intrusion Detection System helps in identifying the security concerns of a network, by monitoring for any anomalies in the network connections. We have proposed a CRF based IDS system using genetic search feature selection algorithm for network security situational awareness to detect any anomalies in the network. The conditional random fields being discriminative models are capable of directly modeling the conditional probabilities rather than joint probabilities there by achieving better classification accuracy. The genetic search feature selection algorithm is capable of identifying the optimal subset among the features based on the best population of features associated with the target class. The proposed system, when trained and tested on the bench mark NSL-KDD dataset exhibited higher accuracy in identifying an attack and also classifying the attack category.

• 정보처리학회논문지B
• /
• 제14B권2호
• /
• pp.135-140
• /
• 2007

특징선택은 문제 영역에서 관찰된 다차원데이터로부터 데이터가 묘사하는 구조를 잘 반영하는 속성을 선택하여 효과적인 실험 데이터를 구성하는 데이터 준비과정이다. 이 과정은 문서분류, 영상인식, 유전자 선택 분야에서의 같은 분류시스템의 성능향상에 중요한 구성요소로서 상관관계 기법, 차원축소 및 상호 정보 처리 등의 통계학이나 정보이론의 접근방법을 중심으로 연구되어왔다. 이와 같은 선택 분야의 연구는 다루는 데이터의 양이 방대해지고 복잡해지면서 더욱 중요시 되고 있다. 본 논문에서는 데이터가 가지는 특성을 반영하면서 새로운 데이터에 대하여 일반화 할 수 있는 특징선택 방법을 제안하고자 한다. 준비된 데이터의 각 속성 데이터에 대하여 퍼지 클러스터 분석에 의하여 최적의 클러스터 정보를 얻고 이를 바탕으로 근접성과 분리성의 경로를 측정하여 그 값에 따라 특징을 선택하는 매카니즘을 제공한다. 제안된 방법을 실세계의 컴퓨터 바이러스 분류에 적용하여 기존의 대비에 의한 휴리스틱 방법에 의해 선택된 데이터를 가지고 분류한 것과 비교하고자 한다. 이를 통하여 주어진 특징에 시연을 부여할 수 있고 효과적으로 특징을 선택하여 시스템의 성능을 향상 시킬 수 있음을 확인한다.

• Journal of Forest and Environmental Science
• /
• 제24권3호
• /
• pp.137-142
• /
• 2008

In this paper, we apply a logistic regression model to the data of snow damage on sugi (Cryptomeria japonica) occurred in Toyama prefecture (in Japan) in 2004 for estimating the risk probability. In order to specify the factors effecting snow damage, we apply a model selection procedure determining optimal subset of explanatory variables. In this process we consider the following 3 information criteria, 1) Akaike's information criterion, 2) Baysian information criterion, 3) Bias-corrected Akaike's information criterion. For the selected variables, we give a proper interpretation from the viewpoint of natural disaster.

• Communications for Statistical Applications and Methods
• /
• 제25권5호
• /
• pp.453-470
• /
• 2018

We study how to distinguish the parameters of the sparse autoregressive (AR) process from zero using a non-convex penalized estimation. A class of non-convex penalties are considered that include the smoothly clipped absolute deviation and minimax concave penalties as special examples. We prove that the penalized estimators achieve some standard theoretical properties such as weak and strong oracle properties which have been proved in sparse linear regression framework. The results hold when the maximal order of the AR process increases to infinity and the minimal size of true non-zero parameters decreases toward zero as the sample size increases. Further, we construct a practical method to select tuning parameters using generalized information criterion, of which the minimizer asymptotically recovers the best theoretical non-penalized estimator of the sparse AR process. Simulation studies are given to confirm the theoretical results.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2002년도 ITC-CSCC -3
• /
• pp.1960-1963
• /
• 2002

Transmit diversity schemes we an effective capacity improvement method for down link of wideband code division multiple access (W-CDMA) systems. In this paper, we propose to use transmit antenna subset selection scheme in conjunction with closed loop transmit adaptive array (TxAA). The proposed scheme selects N$\_$s/ optimum antennas among N$\_$${\gamma}$/(>N$\_$s/) transmit antennas in order to maximize diversity gain from selected antennas, and also reduces the cost of RE chains by employing two different types of RF modules fur the selected and the unselected antenna group, respectively. Computer simulation results show performance improvement by the proposed scheme over the conventional TxAA when considering up link control information feedback.

• ETRI Journal
• /
• 제46권2호
• /
• pp.250-262
• /
• 2024

Air-writing recognition is relevant in areas such as natural human-computer interaction, augmented reality, and virtual reality. A trajectory is the most natural way to represent air writing. We analyze the recognition accuracy of words written in air considering five features, namely, writing direction, curvature, trajectory, orthocenter, and ellipsoid, as well as different parameters of a hidden Markov model classifier. Experiments were performed on two representative datasets, whose sample trajectories were collected using a Leap Motion Controller from a fingertip performing air writing. Dataset D₁ contains 840 English words from 21 classes, and dataset D₂ contains 1600 English words from 40 classes. A genetic algorithm was combined with a hidden Markov model classifier to obtain the best subset of features. Combination ftrajectory, orthocenter, writing direction, curvatureg provided the best feature set, achieving recognition accuracies on datasets D₁ and D₂ of 98.81% and 83.58%, respectively.

• Journal of Information Processing Systems
• /
• 제9권1호
• /
• pp.31-40
• /
• 2013

In this paper, a novel method is proposed to build an ensemble of classifiers by using a feature selection schema. The feature selection schema identifies the best feature sets that affect the arrhythmia classification. Firstly, a number of feature subsets are extracted by applying the feature selection schema to the original dataset. Then classification models are built by using the each feature subset. Finally, we combine the classification models by adopting a voting approach to form a classification ensemble. The voting approach in our method involves both classification error rate and feature selection rate to calculate the score of the each classifier in the ensemble. In our method, the feature selection rate depends on the extracting order of the feature subsets. In the experiment, we applied our method to arrhythmia dataset and generated three top disjointed feature sets. We then built three classifiers based on the top-three feature subsets and formed the classifier ensemble by using the voting approach. Our method can improve the classification accuracy in high dimensional dataset. The performance of each classifier and the performance of their ensemble were higher than the performance of the classifier that was based on whole feature space of the dataset. The classification performance was improved and a more stable classification model could be constructed with the proposed approach.

• 한국통신학회논문지
• /
• 제32권10C호
• /
• pp.965-974
• /
• 2007

본 논문에서는 내용 기반 음악 범주 분류 시스템에서 다중 범주를 위한 특징벡터 선택 알고리즘을 제안한다. 제안된 특징벡터 선택 알고리즘은 분리 성능을 측정할 때 가우시안 혼합 모델(Gaussian Mixture Model: GMM)을 기반으로 GMM separation score을 측정함으로써 확률분포 및 분리 성능 추정의 정확도를 높였고, sequential forward selection 방법을 개선하여 이전까지 선택된 특징벡터들이 분리를 잘 하지 못하는 범주들을 기준으로 다음 특징벡터를 선택하는 알고리즘을 제안하여 다중 범주 분류의 성능을 높였다. 제안된 알고리즘의 성능 검증을 위해 음색, 리듬, 피치 등 오디오 신호의 특징을 나타내는 다양한 파라미터를 오디오 신호로부터 추출하여 제안된 특징벡터 선택 알고리즘과 기존의 알고리즘으로 특징벡터를 선택한 후 GMM classifier와 k-NN classifier를 이용하여 분류 성능을 평가하였다. 제안된 특징벡터 선택 알고리즘은 기존 알고리즘에 비하여 3%에서 8% 정도의 분류 성능이 향상된 것을 확인할 수 있었고 특히 낮은 차원의 특징벡터의 분류 실험에서는 분류 정확도 측면에서 5%에서 10% 향상된 좋은 성능을 보였다.