• 한국멀티미디어학회논문지
• /
• 제22권2호
• /
• pp.203-209
• /
• 2019

The performance of an intelligent classifier for detecting malwares added to multimedia contents based on machine learning is highly dependent on the properties of feature set. Especially, in order to determine the malicious code in real time the size of feature set should be as short as possible without reducing the accuracy. In this paper, we introduce an optimal feature selection method to satisfy both high detection rate and the minimum length of feature set against the feature set provided by PEFeatureExtractor well known as a feature extraction tool. For the evaluation of the proposed method, we perform the experiments using Windows Portable Executables 32bits.

• 한국통신학회논문지
• /
• 제21권7호
• /
• pp.1642-1653
• /
• 1996

This paper covers feature extraction for automatic handwritten signature verification. Several major feature selection techniques are investigated from a practical perspective to realise an optimal signature verification system, and customised feature set selection based on set-on-set distance measurement is presented. The experimental results have proved the proposed methods to be efficient, offering considerably improved verification performance compared to conventional methods. Also, they dramatically reduce the processing complexity in the verification system.

• Journal of Computing Science and Engineering
• /
• 제11권4호
• /
• pp.130-141
• /
• 2017

Feature selection is one of the most challenging problems of pattern recognition and data mining. In this paper, a feature selection algorithm based on an improved version of binary differential evolution is proposed. The method simultaneously optimizes two feature selection criteria, namely, set approximation accuracy of rough set theory and relational algebra based derived score, in order to select the most relevant feature subset from an entire feature set. Superiority of the proposed method over other state-of-the-art methods is confirmed by experimental results, which is conducted over seven publicly available benchmark datasets of different characteristics such as a low number of objects with a high number of features, and a high number of objects with a low number of features.

• 인터넷정보학회논문지
• /
• 제14권5호
• /
• pp.49-57
• /
• 2013

본 연구에서는 효율적인 온 라인 문서 자동 분류를 위해 매우 중요한 분류 작업의 전처리 단계인 특징선택을 위한 새로운 방법이 제안된다. 대부분의 기존 특징선택 방법 연구에서는 특징 집합의 모집단이 단일 모집단으로써 한 모집단이 가지는 정보만으로 분류에 적합한 특징들을 선택하여 특징 집합을 구성하였다. 본 연구에서는 단일 모집단에 한하여 수행되는 특징선택 뿐 만 아니라, 다중 모집단을 가지는 혼합 특징 집합에 대해서 특징선택을 함으로써 다양한 정보를 바탕으로 한 특징 집합을 구성하였다. 혼합 특징 집합은 두 종류의 특징 집합으로 구성된다. 즉 각각 문서로부터 추출한 단어로 구성된 원본 특징 집합과 원본 특징 집합으로부터 LSA를 이용하여 새로 생성한 변형 특징 집합이다. 혼합 특징 집합으로부터 필터 방법과 래퍼 방법을 이용한 하이브리드 방식의 특징 선택을 통해 최적의 특징 집합을 찾고, 이를 이용하여 문서 분류 실험을 수행하였다. 다양한 모집단의 특징들의 정보를 모두 고려함으로써 보다 향상된 분류 성능을 보일 것이라고 기대하였고, 인터넷 뉴스 기사를 대상으로 분류 실험한 결과 90% 이상의 향상된 분류성능을 확인하였다. 특히, 재현율과 정밀도 모두 90%이상의 성능을 보였으며, 둘 사이의 편차가 낮은 것을 확인하였다.

• 융합보안논문지
• /
• 제18권5_1호
• /
• pp.59-66
• /
• 2018

기계학습을 이용하는 침입 탐지 시스템의 성능은 특징 집합의 구성과 크기에 크게 좌우된다. 탐지율과 같은 시스템의 탐지 정확도는 특징 집합의 구성에, 학습 및 탐지 시간은 특징 집합의 크기에 의존한다. 따라서 즉각적인 대응이 필수인 침입 탐지 시스템의 실시간 탐지가 가능하도록 하려면, 특징 집합은 크기가 작으면서도 적절한 특징들로 구성하여야 한다. 본 논문은 실시간 탐지를 위한 특징 집합 선택 문제를 해결하기 위해 사용했던 기존의 다목적 유전자 알고리즘에 특징 간의 Pearson 상관계수를 함께 사용하면 탐지율을 거의 낮추지 않으면서도 특징 집합의 크기를 줄일 수 있음을 보인다. 제안한 방법의 성능평가를 위해 NSL_KDD 데이터를 사용하여 10가지 공격 유형과 정상적인 트래픽을 구별하도록 인공신경망을 설계, 구현하여 실험한다.

• 한국정보과학회논문지:소프트웨어및응용
• /
• 제29권9호
• /
• pp.654-668
• /
• 2002

In this paper, we present an empirical study for improving the Korean text chunking based on machine learning and feature set selection approaches. We focus on two issues: the problem of selecting feature set for Korean chunking, and the problem of alleviating the data sparseness. To select a proper feature set, we use a heuristic method of searching through the space of feature sets using the estimated performance from a machine learning algorithm as a measure of "incremental usefulness" of a particular feature set. Besides, for smoothing the data sparseness, we suggest a method of using a general part-of-speech tag set and selective lexical information under the consideration of Korean language characteristics. Experimental results showed that chunk tags and lexical information within a given context window are important features and spacing unit information is less important than others, which are independent on the machine teaming techniques. Furthermore, using the selective lexical information gives not only a smoothing effect but also the reduction of the feature space than using all of lexical information. Korean text chunking based on the memory-based learning and the decision tree learning with the selected feature space showed the performance of precision/recall of 90.99%/92.52%, and 93.39%/93.41% respectively.

• Journal of Information Science Theory and Practice
• /
• 제2권3호
• /
• pp.29-39
• /
• 2014

In recent years, several studies have proposed making use of the Twitter micro-blogging service to track various trends in online media and discussion. In this study, we specifically examine the use of Twitter to track discussions of food safety in the Korean language. Given the irregularity of keyword use in most tweets, we focus on optimistic machine-learning and feature set selection to classify collected tweets. We build the classifier model using Naive Bayes & Naive Bayes Multinomial, Support Vector Machine, and Decision Tree Algorithms, all of which show good performance. To select an optimum feature set, we construct a basic feature set as a standard for performance comparison, so that further test feature sets can be evaluated. Experiments show that precision and F-measure performance are best when using a Naive Bayes Multinomial classifier model with a test feature set defined by extracting Substantive, Predicate, Modifier, and Interjection parts of speech.

• 한국컴퓨터정보학회논문지
• /
• 제26권10호
• /
• pp.19-26
• /
• 2021

딥러닝 기반 악성코드 탐지 및 분류모델의 성능은 특성집합을 어떻게 구성하느냐에 따라 크게 좌우된다. 본 논문에서는 CNN 기반의 안드로이드 악성코드 탐지 시 탐지성능을 극대화할 수 있는 최적의 특성집합(feature set)을 선정하는 방법을 제안한다. 특성집합에 포함될 특성은 기계학습 및 딥러닝에서 특성추출을 위해 널리 사용되는 Chi-Square test 알고리즘을 사용하여 선정하였다. CICANDMAL2017 데이터세트를 대상으로 선정된 36개의 특성을 이용하여 CNN 모델을 학습시킨 후 악성코드 탐지성능을 측정한 결과 이진분류에서는 99.99%, 다중분류에서는 98.55%의 Accuracy를 달성하였다.

• 한국CDE학회논문집
• /
• 제4권2호
• /
• pp.139-152
• /
• 1999

This paper presents a feature-based approach to extracting machining features fro a feature-based design model. In the approach, a design feature to machining feature conversion process incrementally converts each added design feature into a machining feature or a set of machining features. The proposed approach an efficiently handle protrusion features and interacting features since it takes advantage of design feature information, design intent, and functional requirements during feature extraction. Protrusion features cannot be directly mapped into machining features so that the removal volumes surrounding protrusion features are extracted and converted it no machining features. By utilizing feature information as well as geometry information during feature extraction, the proposed approach can easily overcome inherent problems relating to feature recognition such as feature interactions and loss of design intent. In addition, a feature extraction process can be simplified, and a large set of complex part can be handled with ease.

• Management Science and Financial Engineering
• /
• 제18권2호
• /
• pp.13-17
• /
• 2012

Set covering has widely been accepted as a staple tool for feature selection in data mining. We present a generalized version of this classical combinatorial optimization model to make it better suited for the purpose and propose a surrogate relaxation-based procedure for its meta-heuristic solution. Mathematically and also numerically with experiments on 25 set covering instances, we demonstrate the utility of the proposed model and the proposed solution method.