• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2008.04a
• /
• pp.317-320
• /
• 2008

본 논문에서는 Type-1 퍼지 논리 시스템과 Type-2 퍼지 논리 시스템을 설계하고, 불확실한 정보를 갖는 입력 데이터에 대하여 각각의 성능을 비교한다. Type-1 퍼지 논리 시스템은 외부잡음에 민감한 단점을 가지고 있는 반면, Type-2 퍼지 논리 시스템은 불확실한 정보를 잘 표현할 수 있으며 효율적으로 취급한다. 따라서 Type-2 퍼지 논리 시스템을 이용하여 이러한 단점을 극복하고자 2가지의 모델을 설계한다. 첫 번째 모델은 규칙의 전 ${\cdot}$ 후반부가 불확실성을 표현 할 수 없는 Type-1 퍼지 집합으로 구성된 Type-1 퍼지 논리 시스템을 설계한다. 두 번째는 규칙 후반부만 Type-2 퍼지 집합으로 구성한 두가지의 Type-2 퍼지 논리 시스템을 설계한다. 여기서 규칙 전반부의 입력 공간 분할에는 Min-Max 방법의 균등분할을 사용하고, 규칙 후반부 멤버쉽 함수의 중심 결정에는 입자 군집 최적화(Particle Swarm Optimization) 알고리즘을 사용하여 동정한다. 또한 입력 데이터에 인위적으로 가하는 노이즈의 정도에 따른 각각 모델의 성능을 비교한다. 마지막으로 비선형 모델 평가에 주로 사용되는 가스로 시계열 데이터를 제안된 모델에 적용하고, 실험을 통하여 불확실한 정보를 다루기에 Type-1 퍼지 논리 시스템 보다 Type-2 퍼지 논리 시스템이 효율적이라는 것을 보인다.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2008.04a
• /
• pp.325-328
• /
• 2008

본 논문에서는 비선형 모델의 설계를 위해 Type-2 퍼지 논리 집합을 이용하여 불확실성 문제를 다룬다. 퍼지 논리 시스템의 멤버쉽 함수와 규칙의 구조는 불확실성이 존재하는 언어적인 정보 또는 수치적 데이터를 바탕으로 설계된다. 기존의 Type-1 퍼지 논리 시스템은 외부의 노이즈와 같은 불확실성을 효율적으로 취급할 수 없다. 그러나 Type-2 퍼지 논리 시스템은 불확실한 정보까지 멤버쉽 함수로 표현함으로서 불확실성을 효과적으로 다룰 수 있다. 따라서 본 논문에서는 규칙의 전 ${\cdot}$ 후반부가 Type-2 퍼지 집합으로 구성된 Type-2 퍼지 논리 시스템을 설계하고 불확실성의 변화에 대한 비선형 모델의 성능을 비교한다. 여기서 규칙 전반부 멤버쉽 함수의 정점 선택은 C-means 클러스터링 알고리즘을 이용하고, 규칙 후반부 퍼지 집합의 정점 결정에는 입자 군집 최적화(PSO : Particle Swarm Optimization) 알고리즘을 사용한다. 마지막으로, 비선형 모델 평가에 대표적으로 이용되는 가스로 시계열 데이터를 제안된 모델에 적용하고, 입력 데이터에 인위적인 노이즈가 포함되었을 경우 Type-2 퍼지 논리 시스템이 기존의 Type-1 퍼지 논리 시스템보다 우수함을 보인다.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.21 no.5
• /
• pp.569-574
• /
• 2011

In this paper, we introduce the subspace contrasting group identification problem and propose an algorithm to solve the problem. In order to identify contrasting groups, the algorithm first determines two groups of which attribute values are in one of the contrasting ranges specified by the analyst, and searches for the contrasting groups while increasing the dimension of subspaces with an association rule mining strategy. Because the dimension of microarray data is likely to be tens of thousands, it is burdensome to find all contrasting groups over all possible subspaces by query generation. It is very useful in the sense that the proposed method allows to find those contrasting groups without analyst's involvement.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.24 no.6
• /
• pp.603-608
• /
• 2014

In the maritime traffic environment, the near-miss between vessels is the situation approaching on collision course but collision accident is not occurred. In this study, in order to calculate the near-miss between navigating vessels, the discriminating equation using ship bumper theory and vessel position clustering methods are proposed. Applying proposed module to the vessel trajectories of the WANDO waterway, we assessment navigational risk factors of vessel type, navigational speed, meeting situation.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.18 no.4
• /
• pp.566-571
• /
• 2008

In this paper, we propose a footstep planning method of biped robot based on the Particle Swarm Optimization(PSO). We define configuration and locomotion primitives for biped robots in the 2 dimensional workspace. A footstep planning method is designed using learning process of PSO that is initialized with a population of random objects and searches for optima by updating generations. The footstep planner searches for a feasible sequence of locomotion primitives between a starting point and a goal, and generates a path that avoids the obstacles. We design a path optimization algorithm that optimizes the footstep number and planning cost based on the path generated in the PSO learning process. The proposed planning method is verified by simulation examples in cluttered environments.

• Journal of Intelligence and Information Systems
• /
• v.9 no.3
• /
• pp.177-191
• /
• 2003

Recommender systems are a personalized information filtering technology to help customers find the products they would like to purchase. Collaborative filtering is the most successful recommendation technology. Web usage mining and clustering analysis are widely used in the recommendation field. In this paper, we propose several hybrid collaborative filtering-based recommender procedures to address the effect of web usage mining and cluster analysis. Through the experiment with real e-commerce data, it is found that collaborative filtering using web log data can perform recommendation tasks effectively, but using cluster analysis can perform efficiently.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2015.10a
• /
• pp.297-300
• /
• 2015

이 논문에서는 가축관리의 효율성을 높일 수 있는 IoT 기술 기반의 지능형 축산정보 통합관리 시스템 및 구축 방안을 제안한다. 제안한 시스템은 체온, 활동량, 군집도와 같은 개체특성 정보와 온도, 습도, CO2 농도 등의 축사환경 정보를 무선센서 네트워크를 통해 실시간으로 수집한다. 수집된 정보는 가축개체별 생애주기 사양관리 스케쥴과 연동되어 신속한 관리가 필요한 이상상황을 자동으로 인식한다. 또한, 축적된 양질의 축산지식 정보는 허가된 범위 내에서 접종사, 수정사, 수의사, 축산공무원 등 축산종사자에게 공유된다. 제안한 시스템은 대규모 축사에서 수정, 분만, 질병관리 등 사양관리 시기를 자동으로 감지하여 축주에게 알려줄 수 있기 때문에 최소한의 시간과 비용으로 가축관리 효율성과 생산성을 높일 수 있다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2022.11a
• /
• pp.365-367
• /
• 2022

본 논문에서는 서울시를 대상으로 2020년 기준 자치구별 공공체육시설의 개수를 분석하고, 도출된 서비스 지역 적정 개소 수를 기준으로 추가 설치가 필요한 자치구 내 입지를 예측하였다. 기존 공공 체육시설 수와 선행연구의 입지 지표를 활용해 회귀분석을 바탕으로 유의한 입지요인을 도출하고, 이를 변수로 한 k-means 군집화를 통해 자치구별 입지 후보군이 될만한 행정구역상 동을 구분하였다. 이후 선정된 행정구역 내 기준 인구 당 공공체육시설 비율이 같아지도록 공공체육시설 설치 개수를 결정한 다음 각 구역의 중심점으로부터 가까운 동 순으로 공공체육시설의 추가 설치가 필요한 동을 선정하였다.

• Journal of Internet Computing and Services
• /
• v.21 no.2
• /
• pp.81-90
• /
• 2020

The purpose of this study is to study how to apply block-chain technology to prevent data forgery and alteration in the defense sector of AI(Artificial intelligence). AI is a technology for predicting big data by clustering or classifying it by applying various machine learning methodologies, and military powers including the U.S. have reached the completion stage of technology. If data-based AI's data forgery and modulation occurs, the processing process of the data, even if it is perfect, could be the biggest enemy risk factor, and the falsification and modification of the data can be too easy in the form of hacking. Unexpected attacks could occur if data used by weaponized AI is hacked and manipulated by North Korea. Therefore, a technology that prevents data from being falsified and altered is essential for the use of AI. It is expected that data forgery prevention will solve the problem by applying block-chain, a technology that does not damage data, unless more than half of the connected computers agree, even if a single computer is hacked by a distributed storage of encrypted data as a function of seawater.

• Journal of Intelligence and Information Systems
• /
• v.20 no.2
• /
• pp.93-107
• /
• 2014

In this paper, we report what we have observed with regard to user-perspective issue clustering based on multi-layered two-mode network analysis. This work is significant in the context of data collection by companies about customer needs. Most companies have failed to uncover such needs for products or services properly in terms of demographic data such as age, income levels, and purchase history. Because of excessive reliance on limited internal data, most recommendation systems do not provide decision makers with appropriate business information for current business circumstances. However, part of the problem is the increasing regulation of personal data gathering and privacy. This makes demographic or transaction data collection more difficult, and is a significant hurdle for traditional recommendation approaches because these systems demand a great deal of personal data or transaction logs. Our motivation for presenting this paper to academia is our strong belief, and evidence, that most customers' requirements for products can be effectively and efficiently analyzed from unstructured textual data such as Internet news text. In order to derive users' requirements from textual data obtained online, the proposed approach in this paper attempts to construct double two-mode networks, such as a user-news network and news-issue network, and to integrate these into one quasi-network as the input for issue clustering. One of the contributions of this research is the development of a methodology utilizing enormous amounts of unstructured textual data for user-oriented issue clustering by leveraging existing text mining and social network analysis. In order to build multi-layered two-mode networks of news logs, we need some tools such as text mining and topic analysis. We used not only SAS Enterprise Miner 12.1, which provides a text miner module and cluster module for textual data analysis, but also NetMiner 4 for network visualization and analysis. Our approach for user-perspective issue clustering is composed of six main phases: crawling, topic analysis, access pattern analysis, network merging, network conversion, and clustering. In the first phase, we collect visit logs for news sites by crawler. After gathering unstructured news article data, the topic analysis phase extracts issues from each news article in order to build an article-news network. For simplicity, 100 topics are extracted from 13,652 articles. In the third phase, a user-article network is constructed with access patterns derived from web transaction logs. The double two-mode networks are then merged into a quasi-network of user-issue. Finally, in the user-oriented issue-clustering phase, we classify issues through structural equivalence, and compare these with the clustering results from statistical tools and network analysis. An experiment with a large dataset was performed to build a multi-layer two-mode network. After that, we compared the results of issue clustering from SAS with that of network analysis. The experimental dataset was from a web site ranking site, and the biggest portal site in Korea. The sample dataset contains 150 million transaction logs and 13,652 news articles of 5,000 panels over one year. User-article and article-issue networks are constructed and merged into a user-issue quasi-network using Netminer. Our issue-clustering results applied the Partitioning Around Medoids (PAM) algorithm and Multidimensional Scaling (MDS), and are consistent with the results from SAS clustering. In spite of extensive efforts to provide user information with recommendation systems, most projects are successful only when companies have sufficient data about users and transactions. Our proposed methodology, user-perspective issue clustering, can provide practical support to decision-making in companies because it enhances user-related data from unstructured textual data. To overcome the problem of insufficient data from traditional approaches, our methodology infers customers' real interests by utilizing web transaction logs. In addition, we suggest topic analysis and issue clustering as a practical means of issue identification.