• 디지털융복합연구
• /
• 제15권4호
• /
• pp.191-198
• /
• 2017

PHR(Personal Health Record)기반 헬스케어 서비스 플랫폼 지능화를 위해서는 사용자 맞춤형 건강정보 제공서비스가 필요하다. 본 논문에서는 개인 맞춤형 건강정보 추천을 위해서 온톨로지 기반 건강 정보 모델을 제안하였다. 또한 기계학습과 데이터마이닝 기법을 적용한 유사 건강정보 탐사 알고리즘을 설계하였다. 기존의 데이터마이닝 기법중 연관규칙 알고리즘을 확장하여 속성을 기반으로 연관규칙 탐사를 수행하여 지식탐사의 연관성을 높이고 효율적인 탐사시간을 제공할 수 있도록 하였다. 머신러닝의 한 기법인 K근접이웃 알고리즘을 적용하여 사용자 프로파일별 그룹화를 수행하고 유사패턴의 사용자 프로파일을 검색할 수 있도록 하였다. 이는 사용자의 질환과 건강상태에 따른 맞춤형 건강정보 탐사 수행의 효율성을 높인다. 제안된 알고리즘은 개인 맞춤형 헬스케어 서비스 플랫폼에서 추론과정에 적용되어 사용자에게 개인맞춤형건강정보를 추천하는 것을 가능하게 한다. 이는 고령화사회에서 스마트한 자가 건강관리에 활용될 수 있다.

• 정보처리학회논문지:소프트웨어 및 데이터공학
• /
• 제6권4호
• /
• pp.193-202
• /
• 2017

인공지능 기반 지능형 시스템의 개발에는 일반적으로 신뢰성 높은 대규모 지식처리, 지식의 통합과 인간 수준의 이해, 지식기반 인간-기계협업, 전문가 수준의 지능 서비스 등의 효과적 통합이 요구된다. 특히 빅데이터 이해 기반 자가학습형 지식베이스 및 추론 기술 개발을 목표로 하고 있는 과제의 일환으로 개발 중인 WiseKB 통합 플랫폼은 대용량 지식을 저장하여 추론과정을 통한 질의 및 응답이 가능한 대규모 지식 베이스 역할을 수행하며 이를 위하여 지식표현, 자원통합, 지식저장소, 지식베이스, 복합추론, 지식학습 등의 요소기술들의 효과적 통합이 필수적이다. 통합 플랫폼의 효율적 통합을 위해서는 정확한 요구사항 분석이 중요하며, 이는 시스템의 특성을 고려한 적절한 요구사항 분석 방법론의 적용이 필요하다. 대표적인 요구사항 분석 방법인 순차적 방법론과 순환-점진적 방법론은 WiseKB와 같은 시스템의 대규모 복합적 개발 특성을 고려할 때 다양한 요구사항을 체계적으로 파악하기에 한계가 있다. 본 논문에서는 이러한 한계를 개선하고자 순차적 방법과 순환-점진적 방법론을 결합해 각 단점을 보완하고 대규모 복합적 특성을 갖는 시스템의 요구사항 분석을 효율적으로 진행할 수 있는 통합 방법론을 제시하고, 실제 적용을 통해 그 효과를 보인다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권8호
• /
• pp.3917-3941
• /
• 2019

The recent interest in artificial intelligence and machine learning has partly contributed to an interest in the use of such approaches for hyperspectral remote sensing (HRS) imagery classification, as evidenced by the increasing number of deep framework with deep convolutional neural networks (CNN) structures proposed in the literature. In these approaches, the assumption of obtaining high quality deep features by using CNN is not always easy and efficient because of the complex data distribution and the limited sample size. In this paper, conventional handcrafted learning-based multi features based on expert knowledge are introduced as the input of a special designed CNN to improve the pixel description and classification performance of HRS imagery. The introduction of these handcrafted features can reduce the complexity of the original HRS data and reduce the sample requirements by eliminating redundant information and improving the starting point of deep feature training. It also provides some concise and effective features that are not readily available from direct training with CNN. Evaluations using three public HRS datasets demonstrate the utility of our proposed method in HRS classification.

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 2001년도 추계학술대회 학술발표 논문집
• /
• pp.69-73
• /
• 2001

The dynamic characteristics of turning processes are complex, non-linear and time-varying. Consequently, the conventional techniques based on crisp mathematical model may not guarantee surface roughness regulation. This paper presents a fuzzy controller which can regulate surface roughness in milling process using end-mill under varying cutting condition. The fuzzy control rules are established from operator experience and expert knowledge about the process dynamics. regulation which increases productivity and tool life is achieved by adjusting feed-rate according to the variation of cutting conditions. The performance of the proposed controller is evaluated by cutting experiments in the converted CNC milling machine. The result of experiments show that the proposed fuzzy controller has a good surface roughness regulation capability in spite of the variation of cutting conditions.

• 대한산업공학회지
• /
• 제29권3호
• /
• pp.230-238
• /
• 2003

The configuration design of a mechanical product can be efficiently performed when it is based on the functional modeling. There are methodologies, which decompose function from the abstract level to the concrete level and match the functions to physical parts. But it is difficult to carry out an innovative design when the function is matched only to a pre-detined part. This paper describes the configuration design process of a mechanical product with a design expert system, which uses function taxonomy and TRIZ theory. The expert system can propose a functional modeling of a new part. which is not in the existing parts list. The abstraction levels of design knowledge are introduced, which describe the operation of mechanical product in the levels of abstraction. This is the theoretical background of using knowledge of function and TRIZ for configuration design. The expert system is adequate to control this design knowledge. which expresses knowledge of functional modeling, mapping rules between functions and parts, selection of parts, and TRIZ theory. The hierarchy of functions and machine parts are properly expressed by classes and objects in the expert system. A design expert system has been implemented for the configuration design of a train bogie, and a new brake system of the bogie is introduced with the aid of TRIZ's 30 function groups.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 2006년도 춘계학술대회 논문집
• /
• pp.563-564
• /
• 2006

In silicon wafer manufacturing process, the grinding process has been adopted to improve the flatness of water. The grinding of wafer is usually used by the infeed grinding machine. Grinding conditions are spindle speed, feed speed, rotation speed, grinding stone etc. But grinding condition selection and analysis is so difficult in grinding machine. In the intelligent grinding system based on knowledge many researchers have studied expert system, neural network, fuzzy etc. In this paper we deal grinding condition selection method, Taguchi method and Genetic Analysis.

• 지능정보연구
• /
• 제15권4호
• /
• pp.141-158
• /
• 2009

다양한 정보자원들로부터 사용자가 요구하는 맞춤화된 정보를 추출해 내는 것은 더욱 어려워지고 있다. RSS를 비롯하여 개선된 다양한 정보 수집 방법들이 개발되었지만, 여전히 정보가공자인 사람의 도움 없이 필요한 정보들을 수집하여 정리 및 가공하는 작업이 쉽지는 않다. 따라서 본 연구에서는 정보사용자들이 사용 목적에 맞게 정보를 가공하는 부담을 줄여주기 위해 사례기반추론과 온톨로지에 기반한 맞춤형 통합정보생성 프레임워크를 제안한다. 본 프레임워크는 세 단계로 구성된다. 첫째, 수집된 웹 정보를 정보가공의 용이성을 위해 사례로 변환한다. 둘째, 동적 유사도 검색을 통해 수집된 사례들로부터 정보 사용자의 동적 요구사항에 적합한 사례를 검색한다. 셋째, 전 단계에서 추출된 사례를 정보사용자의 요구사항에 보다 적합한 지식으로 가공하기 위해 집중 유사도를 적용한다. 본 프레임워크는 여행자들의 정보수집을 위한 여행정보시스템에 적용되어 그 효과를 입증하였다.

• Journal of Information Processing Systems
• /
• 제5권2호
• /
• pp.69-76
• /
• 2009

Self healing systems are considered as cognation-enabled sub form of fault tolerance system. But our experiments that we report in this paper show that self healing systems can be used for performance optimization, configuration management, access control management and bunch of other functions. The exponential complexity that results from interaction between autonomic systems and users (software and human users) has hindered the deployment and user of intelligent systems for a while now. We show that if that exceptional complexity is converted into self-growing knowledge (policies in our case), can make up for initial development cost of building an intelligent system. In this paper, we report the application of AHSEN (Autonomic Healing-based Self management Engine) to in OKKAM Project infrastructure backbone cluster that mimics the web service based architecture of u-Zone gateway infrastructure. The 'blind' load division on per-request bases is not optimal for distributed and performance hungry infrastructure such as OKKAM. The approach adopted assesses the active threads on the virtual machine and does resource estimates for active processes. The availability of a certain server is represented through worker modules at load server. Our simulation results on the OKKAM infrastructure show that the self healing significantly improves the performance and clearly demarcates the logical ambiguities in contemporary designs of self healing infrastructures proposed for large scale computing infrastructures.

• 한국지능정보시스템학회:학술대회논문집
• /
• 한국지능정보시스템학회 2003년도 춘계학술대회
• /
• pp.400-409
• /
• 2003

Collaborative filtering (CF) recommendation is a knowledge sharing technology for distribution of opinions and facilitating contacts in network society between people with similar interests. The main concerns of the CF algorithm are about prediction accuracy, speed of response time, problem of data sparsity, and scalability. In general, the efforts of improving prediction algorithms and lessening response time are decoupled. We propose a three-step CF recommendation model which is composed of profiling, inferring, and predicting steps while considering prediction accuracy and computing speed simultaneously. This model combines a CF algorithm with two machine learning processes, SOM (Self-Organizing Map) and CBR (Case Based Reasoning) by changing an unsupervised clustering problem into a supervised user preference reasoning problem, which is a novel approach for the CF recommendation field. This paper demonstrates the utility of the CF recommendation based on SOM cluster-indexing CBR with validation against control algorithms through an open dataset of user preference.

• International Journal of Computer Science & Network Security
• /
• 제21권3호
• /
• pp.177-184
• /
• 2021

All over the world, people are affected by many chronic diseases and medical practitioners are working hard to find out the symptoms and remedies for the diseases. Many researchers focus on the feature detection of the disease and trying to get a better health recommendation system. It is necessary to detect the features automatically to provide the most relevant solution for the disease. This research gives the framework of Health Recommendation System (HRS) for identification of relevant and non-redundant features in the dataset for prediction and recommendation of diseases. This system consists of three phases such as Pre-processing, Feature Selection and Performance evaluation. It supports for handling of missing and noisy data using the proposed Imputation of missing data and noise detection based Pre-processing algorithm (IMDNDP). The selection of features from the pre-processed dataset is performed by proposed ensemble-based feature selection using an expert's knowledge (EFS-EK). It is very difficult to detect and monitor the diseases manually and also needs the expertise in the field so that process becomes time consuming. Finally, the prediction and recommendation can be done using Support Vector Machine (SVM) and rule-based approaches.