• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1993년도 하계학술대회 논문집 A
• /
• pp.436-438
• /
• 1993

A rule based fuzzy expert system in self-tune PID controllers is presented in this paper. The rule base. the core of the expert system, is extracted from the Wills' tuning map and the author's knowledge about the implicit relations between PID gains and controlled output response. The overall control system consists of the relay feedback scheme and the expert system, where the one is responsible for initial tuning and the other for subsequent tuning. The PID control system with the proposed fuzzy expert system, shows better convergence rate and control performances than those of a Litt in spite of the fact that the two rule bases are extracted from the same maps provided by Wills.

• 재활치료과학
• /
• 제1권2호
• /
• pp.14-22
• /
• 2012

서론 : 뇌줄중 환자의 상지 운동 결손은 일상생활에서 수행하는 많은 활동에 큰 어려움을 겪게 만든다. 작업치료사들은 다양한 활동을 제공함으로써 일상생활의 참여를 돕기 위해서 노력하지만 학습자의 특성과 환경을 고려하지 않은 채 훈련이 이루어지고 있다. 따라서 본 연구는 운동 학습 이론과 원칙에 대해 알아보고 작업치료 영역에 적용하기 위한 임상적 고려에 대해 제언해보고자 한다. 본론 : 운동학습의 원칙으로는 학습의 단계, 과제의 형태, 실행 또는 연습, 피드백이 있는 학습의 단계에 따라 운동학습의 원칙을 적용할 수 있을 것이다. 초기단계에서는 학습목표와 작업환경에 대한 충분한 이해가 선행되어야 하며 지속적이며 차단된 연습, 전체 과제 연습이 학습을 촉진시킨다. 후기단계에서는 내제적 피드백에 의존하도록 유도하여야 하며 열린과제, 무작위 연습이 수행을 증진시킨다. 결론 : 치료기전으로 제공되는 다양한 운동학습의 원칙을 적용하기 위해서는 치료계획 수립 시에 체계적으로 결정되어야 할 것이다. 작업치료사들은 환자 각각의 평가결과를 바탕으로 환자의 수준에서 수행을 증진시킬 수 있는 효과적인 운동학습의 원칙을 계획하고 치료적 변화를 계속 모니터링하면서 변경시켜나가야 할 것이다.

• 한국지능시스템학회논문지
• /
• 제24권3호
• /
• pp.316-322
• /
• 2014

본 논문은 TV를 이용한 영화시청 환경에서 해당 컨텐트에 대한 시청자의 암묵적 반응과 컨텐트의 멀티모달 피쳐를 실시간으로 측정 및 동기화하여 이를 기반으로 동영상 선호모델을 지속적으로 개선하고 필요시 영화추천을 수행하는 시스템을 제안한다. 제안한 시스템에선 이미지, 소리, 자막 스트림으로부터 실시간 추출되는 저수준 피쳐들과 동기화되어 측정된 얼굴표정, 자세 및 생체신호로부터 해당 동영상이 유발한 시청자의 감정상태를 추정하여 선호모델 학습에 사용한다. 제안한 컨텐트-시청자 연계 추천모델의 일례로서 컨텐트의 오디오 및 자막 정보를 이용하여 시청자의 피부전기활성도로 측정된 arousal반응을 예측할 수 있음을 보인다.

• 한국정보시스템학회지:정보시스템연구
• /
• 제17권3호
• /
• pp.205-233
• /
• 2008

This study suggests the dynamic value chain model, that will be able to not only show changing processes to organization's significant capital by integrating an individual, implicit, and explicit knowledge which affect organizational decision making, but also distinguish the key driver for raising organizational competitive power because it makes possible to analyze sensitivity of performance along with decision making alternatives and policy changes from dynamic view by connecting knowledge management capability, knowledge management activity, and relations with organizational performance with specific strategic map. Recently, a lot of organizations show interest in measuring and evaluating their performance synthetically. In organizations taking knowledge management, they introduce effective value chain model like a dynamic balanced scorecard (DBSC), and therefore they can reflect their knowledge management condition as well as show their changes by checking performance of established vision and strategy periodically. Furthermore, they can ask for their inner members' understanding and participation by communicating with and inspiring their members with awareness that members are one of their group, present a base of benchmarking, and offer significant information for later decision making. The BSC has been a successful framework for measuring an organization's performance in various perspectives through translating an organization's vision and strategy into an interrelated set of key performance indicators and specific actions. The BSC, while having significant strengths over traditional performance measurement methods, however, has its own limitations, due to its static nature, such as overlooking two-way causation between performance indicators and neglecting the impact of delayed feedback flowing from the adoption of new strategies or policy changes. To overcome these limitations, this study employs SD, a methodology for understanding complex systems where dynamic feedback among the interrelated system components significantly impact on the system outcomes. The SD simulation model in the form of DBSC would serve as a useful strategic teaming tool for facilitating an organization's communication process through various scenario analyses as well as predicting the dynamic behavior pattern of their key performance measures over a future time frame. For the demonstration purpose, this study applied the DBSC model to Prototype of Korea manufacturing and service firm.

• 한국시스템다이내믹스연구
• /
• 제5권2호
• /
• pp.67-88
• /
• 2004

This paper attempts to explore the generic pitfalls of the traditional number-crunching methods adopted thus far for the forecast of newly emerging market trends, and present an alternative by introducingsystems thinking to the portable Internet service market as an example, followed by its rationale as a new tool for forecasting and some reasoning about why traditional methods are no longer appropriate. Most adoption models in general to forecast market trends have several limitations due to theirbasic assumptions and prospective. First, they fail to capture dynamic interactions among the factors involved over time, with implicit assumptions of 'unilateral causality' in that each factor contributes as a cause to the effect, i.e., causality runs one way; each factor acts independently the weighting factor of each is fixed, etc. Second, the number-crunching models have no way of taking into account the impact of delayed feedback often caused by introducing new policies and legislative changes on the whole system under investigation. Third, there is not a way to reflect the effect of competition by players.

• 한국지능정보시스템학회:학술대회논문집
• /
• 한국지능정보시스템학회 2001년도 The Pacific Aisan Confrence On Intelligent Systems 2001
• /
• pp.278-282
• /
• 2001

The discovery of tacit knowledge from domain experts is one of the most exciting challenges in today\`s knowledge management. The nature of decision knowledge in determining the quality a firm\`s short-term liquidity is full of abstraction, ambiguity, and incompleteness, and presents a typical tacit knowledge extraction problem. In dealing with knowledge discovery of this nature, we propose a scheme that integrates both knowledge elicitation and knowledge discovery in the knowledge engineering processes. The knowledge elicitation component applies the Verbal Protocol Analysis to establish industrial cases as the basic knowledge data set. The knowledge discovery component then applies fuzzy clustering to the data set to build a fuzzy knowledge based system, which consists of a set of fuzzy rules representing the decision knowledge, and membership functions of each decision factor for verifying linguistic expression in the rules. The experimental results confirm that the proposed scheme can effectively discover the expert\`s tacit knowledge, and works as a feedback mechanism for human experts to fine-tune the conversion processes of converting tacit knowledge into implicit knowledge.

• 지식경영연구
• /
• 제2권1호
• /
• pp.25-44
• /
• 2001

There are two types of organizational knowledge in terms of its creation process: experiential and analytical knowledge. The experiential knowledge is created by repetitive experiences of an individual or team through task execution, while the analytical knowledge is acquired by analyzing accumulated data or information in the organization. The experiential knowledge often remains tacit or implicit in the organization because it is primarily acquired at an individual or team level. Therefore, the issue on the experiential knowledge is to share it actively within the organization. On the other hand, the analytical knowledge is explicit in its nature since it is extracted from data or information. Thus, it is important to guide a systematic creation of the analytical knowledge rather than encourage to share it. The current trend of "knowledge management" mainly focuses on the experiential knowledge - know-how, idea, case, etc - and neglects another important knowledge in the organization. i. e., analytical knowledge. This paper tries to shed a new light on the "knowledge management" arena by introducing rather new perspective in the concept of knowledge. The purpose of this study is to identify the factors affecting the analytical knowledge creation in the organization. We conducted an exploratory case study of three companies with a previously defined research framework and found some critical factors for the analytical knowledge creation. They are "organizational resource", "effectiveness of feedback process", "data source management", and "experimental mind set". Finally, we proposed research model and propositions regarding the analytical knowledge creation in the organization.

• 정보관리학회지
• /
• 제26권1호
• /
• pp.81-105
• /
• 2009

본 연구에서는 웹, IPTV 등의 콘텐츠 유통망에서의 개인화 추천서비스를 위하여 이용자의 콘텐츠 이용행위와 콘텐츠의 위치정보를 활용한 추천방법을 제안하고 있다. 추천방법의 성능향상을 위하여 이용자 및 콘텐츠 프로파일 생성방법과 함께, 이용자의 콘텐츠 이용행위를 암묵적 이용자 피드백으로서 학습과정에 적용하여 이용자 선호도를 분석하였다. 학습과정에서의 이용자 선호도 분석을 위하여 협업여과추천방법 및 내용기반추천 방법을 적용하였다. 또한 보다 정확한 추천을 위한 최종 콘텐츠 추천을 위하여 웹사이트 상의 콘텐츠에 대한 위치정보를 활용한 추천방법을 제안하고 있다. 이를 통하여 보다 효율적이고 정확한 추천 서비스의 제공이 가능할 수 있다.

• Journal of Information Technology Applications and Management
• /
• 제29권3호
• /
• pp.43-55
• /
• 2022

In this study, with the goal of developing a deep learning-based product recommendation model for effective matching of influencers and products, a deep learning model with a collaborative filtering model combined with generalized matrix decomposition(GMF), a collaborative filtering model based on multi-layer perceptron (MLP), and neural collaborative filtering and generalized matrix Factorization (NeuMF), a hybrid model combining GMP and MLP was developed and tested. In particular, we utilize one-class problem free boosting (OCF-B) method to solve the one-class problem that occurs when training is performed only on positive cases using implicit feedback in the deep learning-based collaborative filtering recommendation model. In relation to model selection based on overall experimental results, the MLP model showed highest performance with weighted average precision, weighted average recall, and f1 score were 0.85 in the model (n=3,000, term=15). This study is meaningful in practice as it attempted to commercialize a deep learning-based recommendation system where influencer's promotion data is being accumulated, pactical personalized recommendation service is not yet commercially applied yet.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2008년도 추계학술발표대회
• /
• pp.229-232
• /
• 2008

웹 정보량이 급속히 늘어나면서 원하는 정보를 효율적으로 찾는 검색 기술의 중요성이 커지고 있다. 검색의 정확성을 높이기 위해서는 검색 질의어와 함께 사용자의 환경, 검색 만족도와 같은 다양한 정보가 필요하다. 사용자의 명시적 피드백을 요구하는 것은 거부감을 줄 수 있으므로 사용자의 잠재적 피드백과 연관 검색어 분석을 통해 검색 질의어를 확장하는 연구가 이뤄지고 있다. 그러나 이러한 검색어 확장과 검색 정확성 사이의 상관관계에 대한 분석이 없어 연관 검색어를 정량적으로 평가할 수 없었다. 본 논문에서는 사용자가 검색 질의어를 변경하면서 검색을 반복하는 과정을 사용자의 잠재적 피드백의 하나로 보고 사용자 만족도를 반영하는 페이지 방문 시간과 함께 분석하여 연속적으로 입력된 검색어가 검색 결과 순위와 사용자 만족도에 미치는 영향을 분석하는 방법을 제안하였다. 마우스 클릭 정보 분석을 통하여 사용자의 검색 만족도를 정량화하였고 특정 주제어에서 관련 검색어가 확장되어 가는 과정은 트리 구조로 표현하였다. 이를 통해 하나의 주제어와 관련해 연속적으로 입력된 검색어 집합으로부터 연관검색어를 추출하고 검색 결과의 정확성을 높일 수 있으며 제안된 트리 구조를 다양한 방향으로 분석하여 검색어, 검색 결과, 사용자 만족도, 배경 지식 등 단순 검색어 분석에서는 나타나지 않는 다양한 정보를 얻을 수 있다.