• Review of KIISC
• /
• v.29 no.6
• /
• pp.67-74
• /
• 2019

기계학습 알고리즘을 이용한 다양한 분야에서의 활용사례들이 우리 사회로 점차 확산되어가며, 기계학습을 통해 산출된 모델의 오동작을 유발할 수 있는 공격이 활발히 연구되고 있다. 특히, 한국에서는 딥러닝을 포함해 인공지능을 응용한 융합분야를 국가적 차원에서 추진하고 있으며, 만약 인공지능 모델 자체에서 발생하는 취약점을 보완하지 못하고 사전에 공격을 대비하지 않는다면, 뒤늦은 대응으로 인하여 관련 산업의 활성화가 지연될 수 있는 문제점이 발생할 수도 있다. 본 논문에서는 기계학습 모델에서, 특히 심층 신경망으로 구성되어 있는 모델에서 발생할 수 있는 공격들을 정의하고 연구 동향을 분석, 안전한 기계학습 모델 구성을 위해 필요한 시사점을 제시한다. 구체적으로, 가장 널리 알려진 적대적 사례(adversarial examples) 뿐 아니라, 프라이버시 침해를 유발하는 추론 공격 등이 어떻게 정의되는지 설명한다.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2018.07a
• /
• pp.139-140
• /
• 2018

본 논문에서는 지도 기계 학습 알고리즘 중 하나인 Naive Bayesian (NB) 알고리즘의 데이터 분류 정확도를 향상시키기 위하여 데이터 속성에 Weight를 부여하는 새로운 기법을 제안하였다. 기존에 Decision Tree(DT) 알고리즘의 깊이를 이용하여 Weigth를 부여하는 방법이 제안되었으나, DT를 구축하는데 오버헤드가 크기 때문에 데이터의 실시간 분석이나 자원 제한적인 환경에서의 적용은 어렵다는 단점이 있다. 이를 해결하기 위하여 본 논문에서는 최소한의 데이터를 사용하여 신속하게 DT를 구축하는 Very Fast Decision Tree (VFDT) 알고리즘 기반의 Weight 부여 기법을 제안함으로써 적은 오버헤드로 NB의 정확도를 향상시킨다.

• Journal of KIISE
• /
• v.42 no.2
• /
• pp.227-234
• /
• 2015

This paper proposes a modified MI-SVM algorithm by considering data distribution. The previous MI-SVM algorithm seeks the margin by considering the "most positive" instance in a positive bag. Positive instances included in positive bags are located in a similar area in a feature space. In order to reflect this characteristic of positive instances, the proposed method selects the "most positive" instance by calculating the distance between each instance in the bag and a pivot point that is the intersection point of all positive instances. This paper suggests two ways to select the "most positive" pivot point in the training data. First, the algorithm seeks the "most positive" pivot point along the current predicted parameter, and then selects the nearest instance in the bag as a representative from the pivot point. Second, the algorithm finds the "most positive" pivot point by using a Diverse Density framework. Our experiments on 12 benchmark multi-instance data sets show that the proposed method results in higher performance than the previous MI-SVM algorithm.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.28 no.3
• /
• pp.635-642
• /
• 2018

In the Era of the Fourth Industrial Revolution, we live in huge amounts of software. However, as software increases, software vulnerabilities are also increasing. Therefore, it is important to detect and remove software vulnerabilities. Currently, many researches have been studied to predict and detect software security problems, but it takes a long time to detect and does not have high prediction accuracy. Therefore, in this paper, we describe a method for efficiently predicting software vulnerabilities using machine learning algorithms. In addition, various machine learning algorithms are compared through experiments. Experimental results show that the k-nearest neighbors prediction model has the highest prediction rate.

• The Transactions of the Korea Information Processing Society
• /
• v.5 no.11
• /
• pp.2897-2907
• /
• 1998

The production-rule generation from training examples is a hard problem that has large space and many local optimal solutions. Many learning methods are proposed for production-rule generation and genetic algorithms is an alternative learning method. However, traditional genetic algorithms has been known to have an obstacle in converging at the global solution area and show poor efficiency of production-rules generated. In this paper, we propose a production-rule generating method which uses genetic algorithm learning. By analyzing optimal sub-solutions captured by genetic algorithm learning, our method takes advantage of its schema structure and thus generates relatively small rule set.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.41 no.2
• /
• pp.87-94
• /
• 2004

The feature subset selection is used as a preprocessing step of a teaming algorithm. If collected data are irrelevant or redundant information, we can improve the performance of learning by removing these data before creating of the learning model. The feature subset selection can also reduce the search space and the storage requirement. This paper proposed a new feature subset selection algorithm that is using the heuristic function based on entropy to evaluate the performance of the abstracted feature subset and feature selection. The ACS algorithm was used as a search method. We could decrease a size of learning model and unnecessary calculating time by reducing the dimension of the feature that was used for learning.

• Journal of Digital Contents Society
• /
• v.18 no.6
• /
• pp.1165-1173
• /
• 2017

Deep learning method, which is one of the most famous machine learning algorithms, has proven its applicability in various applications and is widely used in digital signal processing. However, it is difficult to apply deep learning technology to IoT devices with limited CPU performance and memory capacity, because a large number of training samples requires a lot of memory and computation time. In particular, if the Arduino with a very small memory capacity of 2K to 8K, is used, there are many limitations in implementing the algorithm. In this paper, we propose a method to optimize the ELM algorithm, which is proved to be accurate and efficient in various fields, on Arduino board. Experiments have shown that multi-class learning is possible up to 15-dimensional data on Arduino UNO with memory capacity of 2KB and possible up to 42-dimensional data on Arduino MEGA with memory capacity of 8KB. To evaluate the experiment, we proved the effectiveness of the proposed algorithm using the data sets generated using gaussian mixture modeling and the public UCI data sets.

• Journal of KIISE
• /
• v.41 no.11
• /
• pp.911-920
• /
• 2014

In this paper, we introduce an automatic scoring system for middle-school level writing test based on using machine learning techniques. We discuss overall process and features for building an automatic English writing scoring system. A "concept answer" which represents an abstract meaning of text is newly introduced in order to evaluate the elaboration of a student's answer. In this work, multiple machine learning algorithms are adopted for scoring English writings. We suggest a decision process "optimal combination" which optimally combines multiple outputs of machine learning algorithms and generates a final single output in order to improve the performance of the automatic scoring. By experiments with actual test data, we evaluate the performance of overall automated English writing scoring system.

• Asia-pacific Journal of Multimedia Services Convergent with Art, Humanities, and Sociology
• /
• v.8 no.3
• /
• pp.57-67
• /
• 2018

The ensemble is a unified approach used for getting better performance by using multiple algorithms in machine learning. In this paper, we introduce boosting and bagging, which have been widely used in ensemble techniques, and design a method using support vector regression, radial basis function network, Gaussian process, and multilayer perceptron. In addition, our experiment was performed by adding a recurrent neural network and MOHID numerical model. The drifter data used for our experimental verification consist of 683 observations in seven regions. The performance of our ensemble technique is verified by comparison with four algorithms each. As verification, mean absolute error was adapted. The presented methods are based on ensemble models using bagging, boosting, and machine learning. The error rate was calculated by assigning the equal weight value and different weight value to each unit model in ensemble. The ensemble model using machine learning showed 61.7% improvement compared to the average of four machine learning technique.

• Journal of KIISE:Software and Applications
• /
• v.28 no.5
• /
• pp.422-430
• /
• 2001

웹 에이전트는 사용자가 웹을 브라우징하는 행위를 모니터하여 사용자의 관심 정보를 학습하고 사용자가 필요로 하는 웹 상의 정보를 자동 제공하는 지능형 시스템이다. 웹 에이전트가 사용자의 선호도를 학습하기 위해서는 귀납적 기계학습을 수행하는데, 이때 학습의 효율을 높이기 위해서는 사용자가 관심있어하는 문서들을 유사한 문서들로 클러스터링하여 학습 시스템에 제공하여야 한다. 본 논문에서는 웹 에이전트의 학습 시스템에 입력되는 학습대상 문서들을 보다 정확하고 효율적으로 클러스터링하여 제공하기 위해서 Top-down 방식과 Bottom-up 방식을 통합 적용한 통합방식 문서 클러스터링과 초기 클러스터 생성을 위한 평가함수를 제시한다. Top-down 방식으로는 개념적 클러스터링 알고리즘인 COBWEB을 적용하고, Bottom-up 방식으로는 교차기반(Intersection-based) 클러스터링 방식인 Etzioni의 클러스터링 알고리즘을 적용하였다.