• Journal of the Korean Institute of Intelligent Systems
• /
• v.14 no.7
• /
• pp.847-851
• /
• 2004

In the original mixture density network(MDN), which was introduced by Bishop and Nabney, the parameters of the conditional probability density function are represented by the output vector of a single multi-layer perceptron. Among the recent modification of the MDNs, there is the so-called modified mixture density network, in which each of the priors, conditional means, and covariances is represented via an independent multi-layer perceptron. In this paper, we consider a further simplification of the modified MDN, in which the conditional means are linear with respect to the input variable together with the development of the MATLAB program for the simplification. In this paper, we first briefly review the original mixture density network, then we also review the modified mixture density network in which independent multi-layer perceptrons play an important role in the learning for the parameters of the conditional probability, and finally present a further modification so that the conditional means are linear in the input. The applicability of the presented method is shown via an illustrative simulation example.

• Journal of KIISE:Software and Applications
• /
• v.33 no.12
• /
• pp.1052-1061
• /
• 2006

This paper presents a method of using virtual examples to improve the classification accuracy for data with nominal attributes. Most of the previous researches on virtual examples focused on data with numeric attributes, and they used domain-specific knowledge to generate useful virtual examples for a particularly targeted learning algorithm. Instead of using domain-specific knowledge, our method samples virtual examples from a naive Bayesian network constructed from the given training set. A sampled example is considered useful if it contributes to the increment of the network's conditional likelihood when added to the training set. A set of useful virtual examples can be collected by repeating this process of sampling followed by evaluation. Experiments have shown that the virtual examples collected this way.can help various learning algorithms to derive classifiers of improved accuracy.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.23 no.5
• /
• pp.412-417
• /
• 2013

Analytical modeling of photovoltaic power systems has been receiving significant attentions in recent years in that it is easy to apply for prediction of its dynamics and fault detection and diagnosis in advanced engineering technologies. This paper presents a novel probabilistic modeling approach for such power systems with a big data sequence. Firstly, we express input/output function of photovoltaic power systems in which solar irradiation and ambient temperature are regarded as input variable and electric power is output variable respectively. Based on this functional relationship, conditional probability for these three random variables(such as irradiation, temperature, and electric power) is mathematically defined and its estimation is accomplished from ratio of numbers of all sample data to numbers of cases related to two input variables, which is efficient in particular for a big data sequence of photovoltaic powers systems. Lastly, we predict the output values from a probabilistic model of photovoltaic power systems by using the expectation theory. Two case studies are carried out for testing reliability of the proposed modeling methodology in this paper.

• The Journal of the Convergence on Culture Technology
• /
• v.10 no.4
• /
• pp.595-600
• /
• 2024

As the use of artificial intelligence technologies such as machine learning increases in research fields that predict learning outcomes or optimize learning pathways, the use of artificial intelligence in education is gradually making progress. This research is gradually evolving into more advanced artificial intelligence methods such as deep learning and reinforcement learning. This study aims to improve the method of predicting future learning performance based on the learner's past learning performance-history data. Therefore, to improve prediction performance, we propose conditional probability applying the Markov Chain method. This method is used to improve the prediction performance of the classifier by allowing the learner to add learning history data to the classification prediction in addition to classification prediction by machine learning. In order to confirm the effectiveness of the proposed method, a total of more than 30 experiments were conducted per algorithm and indicator using empirical data, 'Teaching aid-based early childhood education learning performance data'. As a result of the experiment, higher performance indicators were confirmed in cases using the proposed method than in cases where only the classification algorithm was used in all cases.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2017.01a
• /
• pp.151-154
• /
• 2017

소프트웨어의 개발과정 중 결함을 제거하는 작업인 디버깅을 위해서는 가장 먼저 그 결함의 정확한 위치를 찾아야한다. 이 작업은 많은 시간이 소요되며, 이 시간을 단축시키기 위한 결함 위치 식별 기법들이 소개되었다. 많은 기법들 중 프로그램 커버리지 정보를 학습하여 규칙을 분석하는 인공신경망 기반 선행 연구가 있다. 이를 기반으로 본 논문에서는 문장들 간의 관계를 추가적으로 파악하여 학습 데이터로 사용하는 기법을 제안한다. 특정 문장이 항상 지나는 테스트케이스들 중 나머지 다른 문장들이 지나는 테스트케이스의 비율을 통해 문장들 간의 관계를 나타낸다. 해당 비율을 계산하기 위해 조건부 확률인 베이지안 확률을 사용한다. 베이지안 확률을 통해 얻은 문장들의 관계에 따라 인공신경망 내에서 의심도를 결정하는 웨이트(weight)가 기존 기법과는 다르게 학습된다. 이 차이는 문장들의 의심도를 조정하며, 결과적으로 다중 결함 위치 식별의 정확도를 향상시킨다. 본 논문에서 제안한 기법을 이용하여 실험한 결과, Tarantula 대비 평균 39.8%, 기존 역전파 인공신경망(BPNN) 기반 기법 대비 평균 60.5%의 정확도 향상이 있었음을 확인할 수 있다.

• Journal of Educational Research in Mathematics
• /
• v.15 no.3
• /
• pp.315-334
• /
• 2005

The purpose of this study is to find out the implications on when and how the correlation concept can be taught. we investigate the development time and method of the concept in a statistical perspective those initially have discussed in psychology by Piaget. We first reviewed the 1958 research by Inhelder and Piaget. It was the first one which researched the development of the correlation and has become the foundation of psychological perspective. According to them, the correlation concept needs proportional and probability concept ahead of its development and argued on the coefficient of correlation based on formal and logical position. However, from a statistical perspective, the correlation concept is a part of the distribution concept. So, the level of the correlation concept grows from the comparison of conditional distributions to the conditional probability distribution where the proportional concept and probability concept are applied. As reviewed through the literature, we found that 11-12 years old students in early formal operation stage reasoned about correlation through the comparison of conditional distributions. In our study, we argue that we need to consider the possibility of beginning didactic mediation for correlation concept earlier and the method approaching it in a distribution perspective.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.17 no.3
• /
• pp.316-321
• /
• 2007

We propose a quatitative annotation method for edges in Bayesian networks using given sets of condition-specific data. Bayesian network model has been used widely in various fields to infer probabilistic dependency relationships between entities in target systems. Besides the need for identifying dependency relationships, the annotation of edges in Bayesian networks is required to analyze the meaning of learned Bayesian networks. We assume the training data is composed of several condition-specific data sets. The contribution of each condition-specific data set to each edge in the learned Bayesian network is measured using the ratio of likelihoods between network structures of including and missing the specific edge. The proposed method can be a good approach to make quantitative annotation for learned Bayesian network structures while previous annotation approaches only give qualitative one.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.14 no.7
• /
• pp.878-882
• /
• 2004

In this paper, we propose a hierarchical hybrid neural network for detecting faults of induction motor. Implementing the classifier based on the input and output data, we apply appropriate transform and classification method at each step. In the proposed method, after obtaining the current of state of motor for each period, we transform it by Principle Component Analysis(PCA) to reduce its dimension. Before the training process, we use the conditional Fuzzy C-means(FCM) for obtaining the initial parameters of neural network for more effective learning procedure. From the various simulations, we find that the proposed method shows better performance to detect and diagnosis of induction motor and compare than other methods.

• Annual Conference on Human and Language Technology
• /
• 2022.10a
• /
• pp.277-283
• /
• 2022

사주 풀이란 주어진 사주에 대해서 그에 맞는 해석 글을 생성해주는 작업을 의미한다. 전통적으로 사주 풀이는 온전한 사람의 영역으로 인식되어왔으나, 우리는 본 연구를 통해 사주 풀이 영역도 인공지능으로 대체할 수 있을 것이라는 가능성을 탐구한다. 본 연구에서 우리는 최근 연구되고 있는 자연어 생성분야의 연구들에서 영감을 받아, 사주 유형과 사주 풀이 내에 포함할 명사 키워드를 기반으로 풀이글을 생성하는 인공지능 모델 SaJuTeller를 설계한다. 특히 이전 문맥을 고려하여 풀이글을 생성하는 모델과 단순 사주 유형 및 명사 키워드를 기반으로 풀이글을 생성하는 두가지 모델을 제안하며, 이들 각각의 성능을 분석함으로써 각 모델의 구체적인 활용 방안을 제안한다. 본 연구는 우리가 아는 한 최초의 인공지능 기반 사주풀이 연구이며, 우리는 이를 통해 사주풀이에 요구되는 전문인력의 노력을 경감시킴과 동시에, 다양한 표현을 가진 사주 풀이 글을 생성할 수 있음을 제안한다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2020.05a
• /
• pp.323-326
• /
• 2020

태양광 발전은 이산화탄소 배출로 인한 기후 변화에 대응하는 주요 수단으로 인식되어 수요와 필요성이 급격하게 증가하고 있다. 최적의 태양광 발전 시스템의 운영을 위해서는 정교한 전력수요 및 태양광 발전량 예측 모델이 요구되며, 온도 및 일사량은 태양광 발전량 예측 모델의 필수적인 입력 변수이다. 하지만, 한국 기상청의 동네예보는 일사량에 관한 예측값을 제공하지 않아 정교한 태양광 발전량 예측 모델을 구축하는 것은 어렵다. 이를 위해 일사량 예측 기법에 관한 많은 연구사례가 보고되고 있지만, 다수의 연구들은 충분한 데이터 셋을 이용하여 일사량 예측 모델을 개발하였다. 초기 태양광 발전 시스템 운영을 위해서는 불충분한 데이터 셋을 이용한 예측 모델 개발이 필요하나 이에 대한 사례는 불충분하다. 본 논문은 실제 태양광 발전 시스템에서 수집된 불충분한 데이터 셋을 이용한 단기 일사량 예측 기법을 제안한다. 먼저, 기상청 동네예보의 다양한 기상 요인들을 이용하여 일사량 예측 모델을 위한 입력 변수를 구성한다. 다음으로, 조건부 랜덤 포레스트를 이용하여 일사량 예측 모델을 구성하며, 설명 가능한 일사량 예측뿐만 아니라 더욱더 많은 데이터 셋을 학습하기 위해 시계열 교차검증을 수행한다. 실험 결과, 제안한 기법은 다른 예측 기법들보다 높은 예측 정확도를 보일 뿐만 아니라 설명 가능한 예측 결과를 제시할 수 있음을 보여준다.