• Journal of the military operations research society of Korea
• /
• v.23 no.2
• /
• pp.45-71
• /
• 1997

There has been considerable research recently on uncertainty handling in the fields of artificial intelligence and knowledge-based system. Various numerical and non-numerical methods have been proposed for representing and propagating uncertainty in knowledge-based system. The Bayesian method, the Dempster-Shafer's Evidence Theory, the Certainty Factor model and the Fuzzy Set Theory are most frequently appeared in the knowledge-based system. Each of these four methods views uncertainty from a different perspective and propagates it differently. There is no single method which can handle uncertainty properly in all kinds of knowledge-based systems' domain. Therefore a knowledge-based system will work more effectively when the uncertainty handling method in the system fits to the system's environment. This paper proposed a framework for selecting proper uncertainty handling methods in knowledge-based system with respect to characteristics of problem domain and cognitive styles of experts. A schema with strategic/operational and unstructured/structured classification is employed to differenciate domain. And a schema with systematic/intuitive and preceptive/receptive classification is employed to differenciate experts' cognitive style. The characteristics of uncertainty handling methods are compared with characteristics of problem domains and cognitive styles respectively. Then a proper uncertainty handling method is proposed for each category.

• Smart Structures and Systems
• /
• v.26 no.1
• /
• pp.23-33
• /
• 2020

Majority of the damage in engineering structures is nonlinear. Damage sensitive features (DSFs) extracted by traditional methods from linear time series models cannot effectively handle nonlinearity induced by structural damage. A new DSF is proposed based on vector space cosine similarity (VSCS), which combines K-means cluster analysis and Bayesian discrimination to detect nonlinear structural damage. A reference autoregressive moving average (ARMA) model is built based on measured acceleration data. This study first considers an existing DSF, residual standard deviation (RSD). The DSF is further advanced using the VSCS, and then the advanced VSCS is classified using K-means cluster analysis and Bayes discriminant analysis, respectively. The performance of the proposed approach is then verified using experimental data from a three-story shear building structure, and compared with the results of existing RSD. It is demonstrated that combining the linear ARMA model and the advanced VSCS, with cluster analysis and Bayes discriminant analysis, respectively, is an effective approach for detection of nonlinear damage. This approach improves the reliability and accuracy of the nonlinear damage detection using the linear model and significantly reduces the computational cost. The results indicate that the proposed approach is potential to be a promising damage detection technique.

• Proceedings of the KIEE Conference
• /
• 1997.11a
• /
• pp.600-602
• /
• 1997

In this study, a hybrid neural net consisting of an Adaptive LVQ(ALVQ) algorithm and MLP is proposed to perform speaker identification task. ALVQ is a new learning procedure using adaptively feature vector sequence instead of only one feature vector in training codebooks initialized by LBG algorithm and the optimization criterion of this method is consistent with the speaker classification decision rule. ALVQ aims at providing a compressed, geometrically consistent data representation. It is fit to cover irregular data distributions and computes the distance of the input vector sequence from its nodes. On the other hand, MLP aim at a data representation to fit to discriminate patterns belonging to different classes. It has been shown that MLP nets can approximate Bayesian "optimal" classifiers with high precision, and their output values can be related a-posteriori class probabilities. The different characteristics of these neural models make it possible to devise hybrid neural net systems, consisting of classification modules based on these two different philosophies. The proposed method is compared with LBG algorithm, LVQ algorithm and MLP for performance.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.28B no.10
• /
• pp.831-841
• /
• 1991

A combined model of probabilistic and MLP(multi layer perceptron) model is proposed for the pattern classification of EMG( electromyogram) signals. The MLP model has a problem of not guaranteeing the global minima of error and different quality of approximations to Bayesian probabilities. The probabilistic model is, however, closely related to the estimation error of model parameters and the fidelity of assumptions. A proper combination of these will reduce the effects of the problems and be robust to input variations. Proposed model is able to get the MAP(maximum a posteriori probability) in the probabilistic model by estimating a priori probability distribution using the MLP model adaptively. This method minimize the error probability of the probabilistic model as long as the realization of the MLP model is optimal, and this is a good combination of the probabilistic model and the MLP model for the usage of MLP model reliability. Simulation results show the benefit of the proposed model compared to use the Mlp and the probabilistic model seperately and the average calculation time fro classification is about 50ms in the case of combined motion using an IBM PC 25 MHz 386model.

• Proceedings of the KIEE Conference
• /
• 2004.05a
• /
• pp.126-128
• /
• 2004

As e-mail is becoming increasingly important in every day life activity, mail users spend more and more time organizing and classifying the e-mails they receive into folder. Many existing recommendation systems or text classification are mostly focused on recommending the products for the commercial purposes or web documents. So this study aims to apply these application to e-mail more necessary to users. This paper suggests a dynamic recommendation agent system based on Rule Filtering Component recommending the relevant category to enable users directly to manage the optimum classification when a new e-mail is received as the effective method for E-Mail Management. Moreover we try to improve the accuracy as eliminating the limits of misclassification that can be key in classifying e-mails by category. While the existing Bayesian Learning Algorithm mostly uses the fixed threshold, we prove to improve the satisfaction of users as increasing the accuracy by changing the fixed threshold to the dynamic threshold. We designed main modules by rule filtering component for enhanced scalability and reusability of our system.

• Smart Structures and Systems
• /
• v.29 no.4
• /
• pp.589-598
• /
• 2022

Timber structures are susceptible to structural damages caused by variations in moisture content (MC), inducing severe durability deterioration and safety issues. Therefore, it is of great significance to detect MC levels in timber structures. Compared to current methods for timber MC detection, which are time-consuming and require bulky equipment deployment, Lead Zirconate Titanate (PZT)-enabled stress wave sensing combined with statistic machine learning classification proposed in this paper show the advantage of the portable device and ease of operation. First, stress wave signals from different MC cases are excited and received by PZT sensors through active sensing. Subsequently, two non-baseline features are extracted from these stress wave signals. Finally, these features are fed to a statistic machine learning classifier (i.e., naïve Bayesian classification) to achieve MC detection of timber structures. Numerical simulations validate the feasibility of PZT-enabled sensing to perceive MC variations. Tests referring to five MC cases are conducted to verify the effectiveness of the proposed method. Results present high accuracy for timber MC detection, showing a great potential to conduct rapid and long-term monitoring of the MC level of timber structures in future field applications.

• Proceedings of the Korean Information Science Society Conference
• /
• 2004.10b
• /
• pp.790-792
• /
• 2004

본 논문은 PCM(Possibilistic C-Means) 알고리즘과 베이시안 분류 알고리즘을 통합한 고해상도 위성영상의 효과적인 분류방법을 제안하였다. 제안된 알고리즘은 학습데이터를 참고로 하여 PCM 알고리즘을 반복적인 과정 없이 수행한다. 각 분류항목별로 분류된 데이터에서 평균내부거리 내부에 해당되는 데이터들을 선정하여 각 항목별 비율을 구한 후 베이시안 분류기법의 사전확률로 적용하여 분류를 수행한다 PCM 알고리즘은 각 데이터와 특정 클러스터와의 거리에 소속도를 부여하는 퍼지 C-Means 알고리즘과 달리 소속도를 각 데이터와 클러스터 중심간의 절대거리에 의존하는 방법으로 퍼지 C-Means 알고리즘이 가지는 상대성 문제를 해결하였다. 제안된 분류 기법을 고해상도 다중분광 데이터인 IKONOS 위성영상에 적용하여 분류를 수행한 후 최대우도 분류기법과 비교한다.

• Proceedings of the Korean Information Science Society Conference
• /
• 1999.10b
• /
• pp.154-156
• /
• 1999

문서 분류는 미리 정의된 두 개 또는 그 이상의 클래스에 새로 생성되는 객체들을 할당하는 방법이다. 문서의 자동 분류에 대한 연구는 오래 전부터 연구되어 왔지만 한국어에 대한 적용 및 연구는 다른 분야에 비해 아직까지 활발히 이루어지지 않고 있다. 본 논문에서는 문서를 자동으로 분류하기 위해 문서의 주제어에 가중치를 부여하고, 부족한 문서의 특징을 보충하기 위하여 말뭉치로부터 주제어들과의 상호정보에 의해 추출된 단어를 사용하여 문서를 표현한 후, 가중치를 부여한 문서의 주제어에 베이지안 분류자를 사용하여 문서분류를 수행한다. 실험은 한국어 정보검색 실험용 데이터 집합인 KTset95 문서 4,414개 중 1,300개의 문서를 학습 집합으로, 1,000개의 문서를 분류에 대한 검증 집합으로 사용하였다. 실험 결과, 순수 베이지안 확률을 사용한 기존의 방법보다 실험 집합과 검증 집합에서 각각 1.92%, 4.3% 향상된 분류 정확도를 얻었다.

• Smart Media Journal
• /
• v.2 no.2
• /
• pp.20-27
• /
• 2013

The concept of social media is top of the agenda for many business executives and decision makers, as well as consultants try to identify ways where companies can make profitable use of applications such as Netflix, Flixster. The social media is playing an increasingly important role as the information sources for customers making product choices etc. With the flourish of Web 2.0 technology, customer reviews are becoming more and more useful and important information resources for people to save their time and energy on purchasing products that they want. This paper proposes the Bayesian Probabilistic Classification algorithm to mine the social media review, and evaluates it by different splits and cross validation mechanism from the real data set. The explored study experimental results show the robustness and effectiveness of proposed approach for mining the social media review.

• Communications for Statistical Applications and Methods
• /
• v.22 no.6
• /
• pp.625-637
• /
• 2015

We develop a Bayesian clustering procedure based on a Dirichlet process prior with cluster specific random effects. Gibbs sampling of a normal mixture of linear mixed regressions with a Dirichlet process was implemented to calculate posterior probabilities when the number of clusters was unknown. Our approach (unlike its counterparts) provides simultaneous partitioning and parameter estimation with the computation of the classification probabilities. A Monte Carlo study of curve estimation results showed that the model was useful for function estimation. We find that the proposed Dirichlet process mixture model with cluster specific random effects detects clusters sensitively by combining vague edges into different clusters. Examples are given to show how these models perform on real data.