http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
전문가 시스템 Knowledge 수정을 위한 Qualitative Reasoning 응용
박상찬,허순영,유성진 한국전문가시스템학회 1995 학술대회 Vol.2 No.1
In this paper, we propose a method to enhance the performance of knowledge based decision support systems of which knowledge is volatile and incomplete by nature in dynamically changing situation, by providing meta knowledge augmented by Qualitative Reasoning (QR) approach. The proposed system intends to overcome the potential problem of completeness of knowledge base. Using the deep meta knowledge incorporated in the 4R module, along with the knowledge extracted from applying inductive learning, the system identifies the ongoing processes of the target business organization and amplifies the effects of each pending process to its financial attribute values. In so doing, the system utilizes QR models to enhance or reveal the patterns which are otherwise less obvious. The enhanced patterns can eventually be used to improve the classification of the data samples. The success factor is hinged on the completeness of the QR process knowledge base. With a more precise description of the processes taking place, the influences of each process will lead the prediction toward a direction that can reflect more of the current trend. The preliminary results are successful and shed light to the smooth introduction of Qualitative Reasoning to the business domain from the physical laboratory application.
LAN환경에서의 CBM기반의 상호 참여형 멀티미디어 시스템인 원격강의 시스템의 개발
정세현 한국전문가시스템학회 1998 학술대회 Vol.1 No.1
In this paper, we propose Telelecture system on CBM(Computer Based Multimedia) based collaborative multimedia which supports special lecture and collaborative lecture with DB, Face-to-face effect. This system is an application developed by SDK(Software Development Kit) of DooRaie(Distributed Objected Oriented multimedia Application Crafting Environment for Collaborative) framework And this system is developed on windows 95 and windows NT. This scenario of this system is limited within only cyber lecture room(LAN), but it is possible to support many applications development in various network(MAN, PSTN, WAN) on DooRae. This system has a smooth interaction by video and audio, multiple session, multiple participation, application sharing, toolbox including ICON and whiteboard. Also this system supports real or non-real type without constraint of the time and space.
지능적인 전략계획체제를 설계하고 구축하기 위한 인공신경망 전문가시스템 접근방법
이건창,한재호 한국전문가시스템학회 1994 학술대회 Vol.2 No.1
This paper presents a neural expert system approach to designing an intelligent strategic planning system. Strategic planning requires a high level of expertise about interpreting the rival's competitive strategies. However, this kind of expertise is very difficult to obtain. In this respect, we propose using a neural expert system for building an intelligent strategic planning system. The main recipe of the proposed neural expert system is a bi-directional inference mechanism which enables us to perform the 'what-if' and/ar 'goal-seeking' analysis. Three strategic planning portfolio models were considered such as BCG matrix, Growth/Gain matrix, GE matrix, and Product/Market Evolution Portfolio matrix. Ta implement our idea, we developed a prototype system on 486--PC. Using the Korean automobile data, we performed massive experiments under the experimentally designed competitive situations. experimental results support our research intention that the neural expert systems approach is useful for performing competitive analyses.
Integration of Expert System and Neural Algorithm for Developing an Optimal Path-Finding System
최종욱,김영수,정윤 한국전문가시스템학회 1994 학술대회 Vol.2 No.1
One of the most important system modules in car navigation system is an optimal path-finding system which enables auto mobile driver to identify possible routes and show the most economical path between content location and destination. The optimal path-finding problem becomes complicated when multiple variables are considered such as physical route length, degree of congestion, traffic capacity of intersections, number of intersections and lanes, and existence of free ways. Therefore, many researchers in various fields (management science, computer science, applied mathematics, production planning, satellite launching) attempted to solve the problem by ignoring many variables far problem simplification, by developing intelligent algorithms, ar by developing high-speed hardware. In this research, an integration of expert system technique and neural network optimization technique was attempted to develop an operational path-finding system. Even though the integration project is still underway, we believe that the system will provide goad results within a reasonable time limit. This belief is based an the test results showing that reasonable optimal path can be found by drastically reducing problem space through applying expert's knowledge and then by optimizing routes in the next level. Even with more than 50 nodes in the city, the system, different from conventional algorithms, finds the best route within one minute, which is a very encouraging result.
데이타베이스 구조설계 전문가시스템에서의 분야 비의존적 사례에 기반을 둔 추론방법 연구
백용기,서정연,김길창 한국전문가시스템학회 1994 학술대회 Vol.1 No.1
We propose a design expert system using the case-based reasoning to design a database schema. Instead of maintaining only domain knowledge for designing a target application database, the suggested system relies on hybrid case-base that consists of a Domain Dependent Case-Base (DDCB) and a Domain Independent Case-Base (DICB). This system is particularly useful for naive users (nonexperts). We demonstrate how the DICB can be used to design a database schema, and can guarantee the schema to be in 3NF (3rd Normal Form). We briefly discuss how the DDCB is organized and can be used to design an application database schema. The details of the system internal architecture, case-base, reasoning rules, and system flow are explained. It is also shown how the system works by using an example in designing for university administration problem.
UNIK-PMA : 사후모형분석에 의한 최적화모형과 규칙기반시스템 통합시스템
이재규,송용욱 한국전문가시스템학회 1994 학술대회 Vol.1 No.1
Post-Model Analysis (PMA) is a framework that unifies an optimization model with a rule-based system. Thus, PMA enables the multi-objective decision modelling that considers both numeric and symbolic objectives and decision variables. UNIK-PMA implements the PMA procedure on the knowledge-assisted optimization modeler UNIK-OPT and a backward chaining rule-based system LJNIK-BWD. Knowledge-based optimization modeler like UNIK-OPT is essential to let the rules interpreted into a part of optimization models.