• Asia pacific journal of information systems
• /
• v.9 no.2
• /
• pp.19-37
• /
• 1999

Computer aided process planning(CAPP) is a key for implementing CIM. It is bridge between CAD and CAM and translates the design information into manufacturing instructions. Generally, manufacturing is an area where intelligent systems will not be able to rely on methods requiring formalized knowledge. Manufacturing lacks a body of knowledge that is specific, formalized, and rigorous, and which can be coded as rules or procedures. Thus expertise in manufacturing is developed over a period of many years. Case-based reasoning(CBR) offers a new approach for developing intelligent system. In the case-based approach the problem solving experience of the experts is encoded in the form of cases. CBR's retrieval process can be divided to two step. The first step is matching step, and the second step is selection step. For selecting base case, new preference heuristics were introduced using similarity concept. Similarity concept has three has three dimensions, i.e. entity similarity, structural similarity, and goal similarity. In this paper, bolt's process planning was selected an application domain. Following the test result, the new preference heuristics were approved as a useful procedure in CAPP.

• Annual Conference on Human and Language Technology
• /
• 2010.10a
• /
• pp.100-104
• /
• 2010

개체명 인식은 정보 추출의 한 단계로서 정보검색 분야 뿐 아니라 질의응답과 요약 분야에서 매우 유용하게 사용되고 있다. 본 논문에서는 structural Support Vector Machines(structural SVMs) 및 수정된 Pegasos 알고리즘을 이용한 한국어 개체명 인식 시스템에 대하여 기술하고 기존의 Conditional Random Fields(CRFs)를 이용한 시스템과의 성능을 비교한다. 실험결과 structural SVMs과 수정된 Pegasos 알고리즘이 기존의 CRFs 보다 높은 성능을 보였고(신뢰도 99%에서 통계적으로 유의함), structural SVMs과 수정된 Pegasos 알고리즘의 성능은 큰 차이가 없음(통계적으로 유의하지 않음)을 알 수 있었다. 특히 본 논문에서 제안하는 수정된 Pegasos 알고리즘을 이용한 경우 CRFs를 이용한 시스템보다 높은 성능 (TV 도메인 F1=85.43, 스포츠 도메인 F1=86.79)을 유지하면서 학습 시간은 4%로 줄일 수 있었다.

• Archives of Plastic Surgery
• /
• v.49 no.2
• /
• pp.141-149
• /
• 2022

Breast implant-associated anaplastic large cell lymphoma (BIA-ALCL) is a recently spotlighted T-cell origin non-Hodgkin's lymphoma with an increasing incidence of over 800 cases and 33 deaths reported worldwide. Development of BIA-ALCL is likely a complex process involving many factors, such as the textured implant surface, bacterial biofilm growth, immune response, and patient genetics. As the incidence of BIA-ALCL is expected to increase, it is important for all surgeons and physicians to be aware of this disease entity and acquire thorough knowledge of current evidence-based guidelines and recommendations. Early detection, accurate diagnosis, and appropriate treatment are the foundations of current care.

• Annual Conference on Human and Language Technology
• /
• 2023.10a
• /
• pp.505-509
• /
• 2023

개체명 인식은 주어진 텍스트에서 특정 유형의 개체들을 식별하고 추출하는 작업이다. 일반적인 딥러닝 기반 개체명 인식은 사전에 개체명들을 모두 정의한 뒤 모델을 학습한다. 하지만 실제 학습 환경에서는 지속적으로 새로운 개체명이 등장할 수 있을뿐더러 기존 개체명을 학습한 데이터가 접근이 불가할 수 있다. 또한, 새로 모델을 학습하기 위해 새로운 데이터에 기존 개체명을 수동 태깅하기엔 많은 시간과 비용이 든다. 해결 방안으로 여러 방법론이 제시되었지만 새로운 개체명을 학습하는 과정에서 기존 개체명 지식에 대한 망각 현상이 나타났다. 본 논문에서는 지식증류를 활용한 지속학습이 한국어 개체명 인식에서 기존 지식에 대한 망각을 줄이고 새로운 지식을 학습하는데 효과적임을 보인다. 국립국어원에서 제공한 개체명 인식 데이터로 실험과 평가를 진행하여 성능의 우수성을 보인다.

• ETRI Journal
• /
• v.46 no.1
• /
• pp.35-47
• /
• 2024

This study introduces CR-M-SpanBERT, a coreference resolution (CR) model that utilizes multiple embedding-based span bidirectional encoder representations from transformers, for antecedent recognition in natural language (NL) text. Information extraction studies aimed to extract knowledge from NL text autonomously and cost-effectively. However, the extracted information may not represent knowledge accurately owing to the presence of ambiguous entities. Therefore, we propose a CR model that identifies mentions referring to the same entity in NL text. In the case of CR, it is necessary to understand both the syntax and semantics of the NL text simultaneously. Therefore, multiple embeddings are generated for CR, which can include syntactic and semantic information for each word. We evaluate the effectiveness of CR-M-SpanBERT by comparing it to a model that uses SpanBERT as the language model in CR studies. The results demonstrate that our proposed deep neural network model achieves high-recognition accuracy for extracting antecedents from NL text. Additionally, it requires fewer epochs to achieve an average F1 accuracy greater than 75% compared with the conventional SpanBERT approach.

• Journal of Distribution Science
• /
• v.22 no.4
• /
• pp.47-58
• /
• 2024

Purpose: Achieving organizational effectiveness is the ultimate goal that every business entity or institution targets. To achieve this, organizations need to consider various factors that have an impact on their performance. This article analyzes the distribution influence of six main elements that have a central role in shaping sustainable organizational effectiveness, which are organizational culture, job satisfaction, interpersonal communication, talent management, knowledge management, and information technology. Research Design Data and Methodology: This research uses a quantitative approach, focusing on manufacturing companies located in Surabaya as the main object, involving twenty manufacturing companies as research targets, and 10 employees in each company. The sample selection process was carried out through the application of random sampling techniques. The analysis in this research uses the multiple linear regression method and uses SPSS version 26 software. Results: Distribution of six major factors used in this research are related to each other and contribute significantly to overall organizational effectiveness. Conclusion: Organizations that can combine the distribution of a positive culture, prioritize employee satisfaction, encourage effective communication, manage talent and knowledge efficiently, and utilize information technology wisely will have greater potential to achieve their goals and survive in the intensely competitive business environment.

• The Journal of Society for e-Business Studies
• /
• v.9 no.3
• /
• pp.243-258
• /
• 2004

Building a domain-suited ontology, as a means to implement the Semantic Web, is widely believed to offer users the benefit of exploiting the semantic knowledge constrained in the application. Electronic Catalog, shortly e-Catalog, manages the information about the goods or conditions play an important role in e-commerce domain. Consequently, semantically enriched yet precise information by the ontology may elaborate the business transactions. In this paper, we analyze the semantic relationships embodied within the catalog domain, as the first step towards the ontological modeling of e-catalog. Exploring ontology should leverage not only the representation of semantic knowledge but also provide the inferencing capability for the model. We employ the EER(extended Entity Relationships) for the basic model. Each modeling construct can be directly translated by DL(Description Logics). Semantic constraints that can be hardly represented in EER are directly modeled in DL.

• Journal of KIISE:Databases
• /
• v.30 no.6
• /
• pp.573-584
• /
• 2003

Recent massive data generation by genomics and proteomics requires bioinformatic tools to extract the biological meaning from the massive results. Here we introduce ROSPath, a database system to deal with information on reactive oxygen species (ROS)-mediated cell signaling pathways. It provides a structured repository for handling pathway related data and tools for querying, displaying, and analyzing pathways. ROSPath data model provides the extensibility for representing incomplete knowledge and the accessibility for linking the existing biochemical databases via the Internet. For flexibility and efficient retrieval, hierarchically structured data model is defined by using the object-oriented model. There are two major data types in ROSPath data model: ‘bio entity’ and ‘interaction’. Bio entity represents a single biochemical entity: a protein or protein state involved in ROS cell-signaling pathways. Interaction, characterized by a list of inputs and outputs, describes various types of relationship among bio entities. Typical interactions are protein state transitions, chemical reactions, and protein-protein interactions. A complex network can be constructed from ROSPath data model and thus provides a foundation for describing and analyzing various biochemical processes.

• Journal of the Korean Society for Precision Engineering
• /
• v.11 no.4
• /
• pp.77-87
• /
• 1994

This paper deals with an automated forming sequence design system by which designers can determine desirable operation sequences even if they have little experience in the design of cold forging process. The forming sequence design in the cold forging is very important and requires many kinds of technical and empirical knowledge. They system isproposed, which generates forming sequence plans for the multistage cold forging of axisymmtrical solid products. Since the process of metal forming can be considered as a transformation of geometry, treatment of the geometry of the product is a key in planning process. To recognize the geometry of the product section, section entity representation and primitive geometries were used. Section entity representation can be used for the calculation of maximum diameter, maximum height, and volume. Forming sequence for the part can be determined by means of primitive geometries such as cylinder, cone, convex, and concave. By utilizing this geometrical characteristics (diameter, height, and radius), the product geometry is expressed by a list of the priitive geometries. Accordingly the forming sequence design is formulated as the search problem which starts with a billet geometry and finishes with a given product one. Using the developed system, the sequence drawing with all dimensions, which includes the proper sequence of operations for the part, is generated under the environment of AutoCAD. Based on the results of forming sequence, process variables(strain, punch pressure, die inner pressure, and forming load) are determined.

• Journal of the Korean Society for Library and Information Science
• /
• v.33 no.4
• /
• pp.133-156
• /
• 1999

The purpose of this paper is to show how conceptual data modeling can enhance current information retrieval (IR) systems. The conceptual database design provides for: 1) data mining capability to discover new knowledge based on the relationships between entities, and 2) integrating current separate databases into one IR system (e.g., integrating ISI Citation, a thesaurus, and bibliographic databases into one retrieval system) . Further, as new user requirements are unfolded, modifications of IR systems based on conceptual data modeling will be much easier to make than they were in the current IR systems because conceptual modeling facilitates flexible modifications. The enhanced Entity-Relationship (ER) model was employed in this study to develop conceptual schemas of IR data.