• Journal of KIISE:Computing Practices and Letters
• /
• v.11 no.3
• /
• pp.275-283
• /
• 2005

Ontologies are the core components of the Semantic Web and knowledge-based systems. But it is difficult to find useful ontologies in actual domains. In order to build useful domain ontology, the conceptualization of the domain knowledge by knowledge experts of the specific domain and the specification of conceptualized knowledge with formal languages by ontology designers are required. In addition, structured and detailed guidelines and methods should be provided to be shared by the development team members. However, existing ontology building methodologies define and describe the skeletal structure of the whole building process at the top-layer. We build a useful academic ontology that is based on the conceptual knowledge structure in the domain of philosophy, and propose a detailed methodology to build a text ontology based on Topic Maps. Our methodology consists of two phases, ontology modelling and ontology implementation. We implement a philosophy knowledge portal to support retrieving and navigating of the philosophy knowledge.

• International Journal of CAD/CAM
• /
• v.5 no.1
• /
• pp.99-109
• /
• 2005

This paper introduces an approach to an ontology-based multi-level knowledge framework for a knowledge management system for discrete-product development. Participants in a product life cycle want to share comprehensive product knowledge without any ambiguity and heterogeneity. However, previous knowledge management approaches are limited in providing those aspects: therefore, we suggest an ontology-based multi-level knowledge framework (OBMKF). The bottom level, the axiom, specifies the semantics of concepts and relations of knowledge so ambiguity can be alleviated. The middle level is a product development knowledge map; it defines the concepts and the relations of the product domain knowledge and guides the engineer to process their engineering decisions. The middle level is then classified further into more detailed levels, such as generic product level, specific product level, product version level, and manufactured item level, according to the various viewpoints. The top level is specialized knowledge for a specific domain that gives the solution of a specific task or problem. It is classified into three knowledge types: expert knowledge, engineering function knowledge, and data-analysis-based knowledge. This proposed framework is based on ontology to accommodate a comprehensive range of knowledge and is represented with first-order logic to maintain a uniform representation.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.3 no.1
• /
• pp.76-89
• /
• 1993

The evaluation of the Fuzzy Expected Value (FEV) as a typical value requires complete knowledge about the domain of the evaluation, and the distribution of the population in that domain [1]. Since in many situations it is not possible to gather complete knowledge regarding the domain, it is necessary to relax some of the restrictions involving the evaluation of FEV. In this paper we discuss solutions to this problem by using the concept of the Fuzzy Expected Interval (FEI).

• Journal of the Korean Society for information Management
• /
• v.6 no.1
• /
• pp.93-117
• /
• 1989

The objective of this study is to design an automatic abstracting system through the analysis of natural language texts. For this purpose a knowledge-based system operating on the basis of domain knowledge was developed. The procedure of generating an abstract consists of three steps: (1) A knowledge-base containing domain knowledge necessary to understand a text is constructed using frame and semantic network structures,and preliminary abstracts are prepared for various cases. (2) Input text is analysed on the basis of domain knowledge in order to extract information filling slots of the abstract with. (3) A Preliminary abstract corresponding to the input text is called and filled with the information, completing the abstract.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.19 no.1
• /
• pp.17-40
• /
• 1994

The objectives of this paper are to apply fuzzy cognitive map (FCM)- related techniques to (1) extract causal knowledge from a specific problem-domain and (2) perform a series of causal analysis in complicated decision making area. We propose a set operation-based augmentation (SOBA) algorithm to combine multiple FCMs developed by multiple experts. Based on the SOBA knowledge acquisition algorithm, we can obtain a causal knowledge base fairly representing multiple experts' knowledge about a problem domain. The causal knowledge base built by SOBA algorithm can be described as a matrix form, guaranteeing mathematically compact operation compared with a production (if-then) knowledge base. We applied out method to stock market analysis problem whichis a typical of highly unstructured problems in OR/MS fields.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 1998.10a
• /
• pp.228-231
• /
• 1998

Artificial Intelligence literatures have recognized that stock market is a highly unstructured and complex domain so that it is difficult to find knowledge that belongs to that domain. This paper demonstrates that the proposed QCOM can derive global knowledge about stock market on the basis of a set of local knowledge and express it as a digraph representation. In addition, inference mechanism using quantitative causal reasoning can describe the qualitative and quantitative effects of exogenous variables on stock market.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.30B no.8
• /
• pp.41-53
• /
• 1993

The tasks which detect and track moving objects, by analyzing dynamic images taken at a constant time interval, are essential in various applications. This paper suggests how to utilize domain-specific knowledge and motional knowledge for detecting and tracking moving objects. That is, The trajectory information of a moving object is to be used for generating hypotheses on expected motion and expected position of moving objects, and the domain-specific knowledge is to be used for verifying the generated hypotheses.

• Journal of Internet Computing and Services
• /
• v.7 no.4
• /
• pp.67-77
• /
• 2006

Wrappers play an important role in extracting specified information from various sources. Wrapper rules by which information is extracted are often created from the domain-specific knowledge. Domain-specific knowledge helps recognizing the meaning the text representing various entities and values and detecting their formats However, such domain knowledge becomes powerless when value-representing data are not labeled with appropriate textual descriptions or there is nothing but a hyper link when certain text labels or values are expected. In order to alleviate these problems, we propose a probabilistic method for recognizing the entity type, i.e. generating wrapper rules, when there is no label associated with value-representing text. In addition, we have devised a method for using the information reachable by following hyperlinks when textual data are not immediately available on the target web page. Our experimental work shows that the proposed methods help increasing precision of the resulting wrapper, particularly extracting the title information, the most important entity on a web page. The proposed methods can be useful in making a more efficient and correct information extraction system for various sources of information without user intervention.

• The Journal of Korea Robotics Society
• /
• v.18 no.3
• /
• pp.346-351
• /
• 2023

With the release of numerous open driving datasets, the demand for domain adaptation in perception tasks has increased, particularly when transferring knowledge from rich datasets to novel domains. However, it is difficult to solve the change 1) in the sensor domain caused by heterogeneous LiDAR sensors and 2) in the environmental domain caused by different environmental factors. We overcome domain differences in the semi-supervised setting with 3-stage model parameter training. First, we pre-train the model with the source dataset with object scaling based on statistics of the object size. Then we fine-tine the partially frozen model weights with copy-and-paste augmentation. The 3D points in the box labels are copied from one scene and pasted to the other scenes. Finally, we use the knowledge distillation method to update the student network with a moving average from the teacher network along with a self-training method with pseudo labels. Test-Time Augmentation with varying z values is employed to predict the final results. Our method achieved 3rd place in ECCV 2022 workshop on the 3D Perception for Autonomous Driving challenge.

• Proceedings of the Technology Innovation Conference
• /
• 1999.06a
• /
• pp.98-118
• /
• 1999

This study deals with the knowledge acquisition activities along the growth stages of Korean ventures. This issue is very important in the three reasons. First, the target of the study-new ventures- is a pending issue and can be regarded as the engine of innovation in the Korean economy. Second, venture activities is so dynamic compared to those of the established companies and the study reflects its dynamic features. Third, the knowledge is becoming more important one among various resources, and knowledge management can be a timely issue. The main research questions are as follows : - How does the degree of knowledge domain, vary along the growth stages\ulcorner - Which knowledge domains are more influential on the performance along growth stages\ulcorner Major findings of the study are as follow: First, technological knowledge acquisition effort are most intensive at the start-up stage, while the management knowledge efforts are active at the growth stage. The degree of market knowledge acquisition efforts is almost the same along the stages. Second, the important knowledge domain, which influences on the performance, varies along the stages. The acquisition effort for product technology knowledge is more influential on the sales growth rate and has a negative effect on the return on assets at the start-up stage, while the management knowledge about administration is more influential on the return on assets at the growth stage. Finally the academic contributions and managerial implications of the study are presented and the future research directions are also suggested.