• 제어로봇시스템학회:학술대회논문집
• /
• 2005.06a
• /
• pp.714-719
• /
• 2005

In this paper, the informative space is proposed to implant ubiquitous functions into physical spaces. We merge physical and virtual spaces through the space structurization using an RFID system, and solve the space localization and mapping problem for a robot to navigate through the distribution and synthesis of information and knowledge. To distribute knowledge flexibly and reliably to changing environment and also to develop a system which allows a robot to invoke and merge the distributed knowledge more freely, we employ a novel approach of knowledge management based on Web services. The proposed method is verified by building a physical space with two kinds of RFID tags and a virtual space with knowledge database based on Web services.

• Proceedings of the IEEK Conference
• /
• summer
• /
• pp.87-91
• /
• 2004

This research presents distributed database and knowledge-base modeling approach for intelligent product design system. The product design information in this study is described by a collection of rules and design knowledge that are utilized according to the product development procedures. In this work, a network-based architecture has been developed to enable dispersed designers to simultaneously accomplish remote design tasks. A client/server communication diagram has also been proposed to facilitate consistent primary information modeling for multi-user access and reuse of designed results. An intelligent product design system has been studied with the concepts of distributed database and network-based architecture in order to support concurrent engineering design and automatic design part assembly. The system provides the capability of composing new designs from proper design elements stored in the database and knowledge-base. The distributed intelligent product design is applied to the design of an automobile part as an example.

• Management & Information Systems Review
• /
• v.3
• /
• pp.179-199
• /
• 1999

Case-Based Reasoning(CBR) offers a new approach for developing knowledge based systems. In case-based approach the problem solving experience of the domain expert is encoded in the form of cases. CBR has successfully been applied to many kinds of problems such as design, planning, diagnosis and forecasting. In this paper, CBR was applied for forecasting default risk. The applied result was successful in spite of the small casebase. Generally, CBR requires large casebase. So, if the number of data was large, the result was better. But in this paper, what financial variable was more forecastable was not tested. Next, this should be tested.

• Annual Conference on Human and Language Technology
• /
• 2018.10a
• /
• pp.652-657
• /
• 2018

With the rapid development of e-commerce, many customers can now express their opinion on various kinds of product at discussion groups, merchant sites, social networks, etc. Discerning a consensus opinion about a product sold online is difficult due to more and more reviews become available on the internet. Opinion Mining, also known as Sentiment analysis, is the task of automatically detecting and understanding the sentimental expressions about a product from customer textual reviews. Recently, researchers have proposed various approaches for evaluation in sentiment mining by applying several techniques for document, sentence and aspect level. Aspect-based sentiment analysis is getting widely interesting of researchers; however, more complex algorithms are needed to address this issue precisely with larger corpora. This paper introduces an approach of knowledge representation for the task of analyzing product aspect rating. We focus on how to form the nature of sentiment representation from textual opinion by utilizing the representation learning methods which include word embedding and compositional vector models. Our experiment is performed on a dataset of reviews from electronic domain and the obtained result show that the proposed system achieved outstanding methods in previous studies.

• Journal of Information Processing Systems
• /
• v.13 no.4
• /
• pp.716-730
• /
• 2017

In this paper, we propose a transliteration approach based on semantic information (i.e., language origin and gender) which are automatically learnt from the person name, aiming to transliterate the person name of Uyghur into Chinese. The proposed approach integrates semantic scores (i.e., performance on language origin and gender detection) with general transliteration model and generates the semantic knowledge-based model which can produce the best candidate transliteration results. In the experiment, we use the datasets which contain the person names of different language origins: Uyghur and Chinese. The results show that the proposed semantic transliteration model substantially outperforms the general transliteration model and greatly improves the mean reciprocal rank (MRR) performance on two datasets, as well as aids in developing more efficient transliteration for named entities.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.8 no.6
• /
• pp.70-77
• /
• 1998

Although many methods of knowledge acquisition has been developed in the exper systems field, such a need form causal knowledge acquisition hs not been stressed relatively. In this respect, this paper is aimed at suggesting a causal knowledge acquisition process, and then investigate the causal knowledge-based inference process. A vehicle for causal knowledge acquisition is FCM (Fuzzy Cognitive Map), a fuzzy signed digraph with causal relationships between concept variables found in a specific application domain. Although FCM has a plenty of generic properties for causal knowledge acquisition, it needs some theoretical improvement for acquiring a more refined causal knowledge. In this sense, we refine fuzzy implications of FCM by proposing fuzzy causal relationship and fuzzy partially causal relationship. To test the validity of our proposed approach, we prototyped a causal knowledge-driven inference engine named CAKES and then experimented with some illustrative examples.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.54 no.5
• /
• pp.35-41
• /
• 2017

In this paper, we analyze the performance of the recently introduced Hint-knowledge distillation (KD) training approach based on the teacher-student framework for knowledge distillation and knowledge transfer. As a deep neural network (DNN) considered in this paper, the deep residual network (ResNet), which is currently regarded as the latest DNN, is used for the teacher-student framework. Therefore, when implementing the Hint-KD training, we investigate the impact on the weight of KD information based on the soften factor in terms of classification accuracy using the widely used open deep learning frameworks, Caffe. As a results, it can be seen that the recognition accuracy of the student model is improved when the fixed value of the KD information is maintained rather than the gradual decrease of the KD information during training.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.2 no.2
• /
• pp.115-119
• /
• 2001

The design and manufacture of injection molded polymeric parts with desired properties is a costly process dominated by empiricism, including the repeated modification of actual tooling. This paper presents an knowledge-based synthesis system which can predict the mechanical performance of a molded product and diagnose the design before the actual mold is machined. The knowledge-based system synergistically combines a rule-based system with CAE programs. Hueristic know]edge of injection molding. flow simulation, and mechanical performance prediction is formalized as rules of an expert consultation system. The expert system interprets the analytical results of the process simulation, predicts the performance, evaluates the design and generates recommendation for optimal design alternatives.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2001.04a
• /
• pp.774-777
• /
• 2001

This paper describes a research work of developing a computer-aided design system of cold forging products. An approach to the system is based on the knowledge-based rules. Knowledge for the system is formulated from plastic theories, experimental results and the empirical knowledge of field experts. The system has been written in AutoCAD with a personal computer and is composed of six modules, which are selection of billet material, input of final product, process planning design, preform modification, insert design, shrink rings design module. Based on knowledge-based rules, the system is designed by considering several factors such as volume constancy, limite of formability by material, preform shape and so on.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2000.05a
• /
• pp.829-832
• /
• 2000

This paper describes a research work of developing a computer-aided design of product with bending and piercing for progressive working. An approach to the CAD system is based on the knowledge-based rules. Knowledge fur the CAD system is formulated from plasticity theories, experimental results and the empirical knowledge of field experts. The system has been written in AutoLISP on the AutoCAD with a personal computer and is composed of three main modules, which are input and shape treatment, flat pattern layout and strip layout module. Based on knowledge-based rules, the system is designed by considering several factors, such as radius and angle of bend, material and thickness of product, complexities of blank geometry and punch profile, bending sequence, and availability of press. Strip layout drawing automatically generated by piercing with punch profiles divided into for external area is simulated in 3-D graphic forms, including bending sequences for the product with piercing and bending. Results obtained using the modules enable the manufacturer of electronic products to be more efficient in this field.