• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제14권8호
• /
• pp.3168-3186
• /
• 2020

Automatic topic essay generation (TEG) is a controllable text generation task that aims to generate informative, diverse, and topic-consistent essays based on multiple topics. To make the generated essays of high quality, a reasonable method should consider both diversity and topic-consistency. Another essential issue is the intrinsic link of the topics, which contributes to making the essays closely surround the semantics of provided topics. However, it remains challenging for TEG to fill the semantic gap between source topic words and target output, and a more powerful model is needed to capture the semantics of given topics. To this end, we propose a pretraining-based contextual self-attention (PC-SAN) model that is built upon the seq2seq framework. For the encoder of our model, we employ a dynamic weight sum of layers from BERT to fully utilize the semantics of topics, which is of great help to fill the gap and improve the quality of the generated essays. In the decoding phase, we also transform the target-side contextual history information into the query layers to alleviate the lack of context in typical self-attention networks (SANs). Experimental results on large-scale paragraph-level Chinese corpora verify that our model is capable of generating diverse, topic-consistent text and essentially makes improvements as compare to strong baselines. Furthermore, extensive analysis validates the effectiveness of contextual embeddings from BERT and contextual history information in SANs.

• 한국산학기술학회:학술대회논문집
• /
• 한국산학기술학회 2003년도 Proceeding
• /
• pp.65-69
• /
• 2003

As complex mathematical models are increasingly adopted for business decision-making, difficulties arise in reusing solvers (i.e., model solving algorithms) against diverse models and data sets and thus the collaboration among users (model/solver builders and decision makers) in multiple departments becomes very difficult. To facilitate the solver reuse, this paper adopts the Web services technologies as the base technologies for linking the solvers to the models, both of which are created on different modeling paradigms and different system platforms, in unified system architecture. Specifically, this paper focuses on designing an ontology that represents the interfacing semantics of the model-solver interactions in a general and standardized form. By referring to the ontology, a model management system (MMS) can autonomously suggest a set of compatible solvers and apply them to individual models even though the decision makers are not knowledgeable enough about all the details of the models and the solvers. Thus, this Web services based MMS would improve the reusability of the solvers by relieving the decision makers from the risk of erroneous application of a solver to syntactically and semantically incompatible models and the burden of considerable understanding of model and solver semantics.

• International Journal of Computer Science & Network Security
• /
• 제21권9호
• /
• pp.247-260
• /
• 2021

Data modeling is a process of developing a model to design and develop a data system that supports an organization's various business processes. A conceptual data model represents a technology-independent specification of structure of data to be stored within a database. The model aims to provide richer expressiveness and incorporate a set of semantics to (a) support the design, control, and integrity parts of the data stored in data management structures and (b) coordinate the viewing of connections and ideas on a database. The described structure of the data is often represented in an entity–relationship (ER) model, which was one of the first data-modeling techniques and is likely to continue to be a popular way of characterizing entity classes, attributes, and relationships. This paper attempts to examine the basic ER modeling notions in order to analyze the concepts to which they refer as well as ways to represent them. In such a mission, we apply a new modeling methodology (thinging machine; TM) to ER in terms of its fundamental building constructs, representation entities, relationships, and attributes. The goal of this venture is to further the understanding of data models and enrich their semantics. Three specific contributions to modeling in this context are incorporated: (a) using the TM model's five generic actions to inject processing in the ER structure; (b) relating the single ontological element of TM modeling (i.e., a thing/machine or thimac) to ER entities and relationships; and (c) proposing a high-level integrated, extended ER model that includes structural and time-oriented notions (e.g., events or behavior).

• 한국멀티미디어학회논문지
• /
• 제21권2호
• /
• pp.158-165
• /
• 2018

Since a large amount of multimedia image has dramatically increased, it is important to search semantically relevant image. Thus, several semantic image annotation methods using RDF(Resource Description Framework) model in mobile environment are introduced. Earlier studies on annotating image semantically focused on both the image tag and the context-aware information such as temporal and spatial data. However, in order to fully express their semantics of image, we need more annotations which are described in RDF model. In this paper, we propose an annotation method inferencing with RDFS entailment rules and user defined rules. Our approach implemented in Moment system shows that it can more fully represent the semantics of image with more annotation triples.

• 한국경영과학회:학술대회논문집
• /
• 대한산업공학회/한국경영과학회 2004년도 춘계공동학술대회 논문집
• /
• pp.710-713
• /
• 2004

There have been numerous research and development efforts to automate business processes or execute business processes automatically. Especially, Web service is considered to be one of the most promising technologies. Existing Web service composition languages such as BPEL (Business Process Execution Language) and BPML (Business Process Modeling Language) do not provide full semantic support for business processes. This paper proposes a 3-layer model that consists of business process layer and service layer which can seamlessly integrates business process semantics with service semantics. The proposed model and architecture will enable effective management of business processes as well-defined corporate assets, services that execute tasks to achieve goals of business processes, and software components that comprise services.

• 한국항만학회지
• /
• 제14권1호
• /
• pp.47-55
• /
• 2000

In order to cope with the changes of container terminal situation in these days, many simulation studies for container terminal have been accomplished. But previous simulation studies using simulation language have limitations in model representation and difficulties in modeling of large scaled container terminal system. To make these problems better, this paper addresses an object-oriented simulation of container terminal system using a DEVS formalism. The DEVS(Discrete Event System Specification) formalism, developed by Zeigler, supports specification of discrete event system in a hierarchical and modular manner. The formalism provides a mathematical basis for studying discrete event systems with better understood and sounder semantics. In a step of system modeling, a DEVS formalism aims at the exact system modeling that has a basis of semantics and utilizing the object-oriented manner can flexibly cope with the changes of system environment. In this study a model is developed and verified through the simulation of some alternatives.

• 한국비블리아학회지
• /
• 제23권1호
• /
• pp.5-26
• /
• 2012

FRSAD과 온톨로지 모형간의 연계는 어떠한 정보구현시스템이나 특정 상황에서도 독립적인 상위 계층의 지식구조를 제안하고, 의미 연결, 지식구조, 주제접근, 상호운용성 등을 확보하기 위한 노력이다. FRSAD 모형을 현실적인 정보환경에 적용하고, 도서관 안팎 어디서나 자유롭게 정보를 확인하고 공유하기 위해서는 지식표현 인코딩 체계가 필요하다. 본 연구에서는 웹 온톨로지 언어인 OWL를 사용하여 FRSAD 모형을 온톨로지 구조로 표현하였으며, 이를 확인하기 위해 Prot$\acute{e}$g$\acute{e}$를 사용하여 FRSAD 온톨로지 모형을 시범적으로 구축하였다.

• 한국산학기술학회논문지
• /
• 제15권1호
• /
• pp.449-459
• /
• 2014

시각적인 클래스 모델로 기 제시된 2+1 View 통합 메타모델은 비정형적인 명세에 기인하여 명확하게 모델의 구문을 표현하지 못하고 있으며, 또한 그 모델의 정확성을 보장할 수 없다. 본 논문은 Z와 Object-Z를 사용해서 2+1 View 통합 메타모델의 구문적 의미를 정형적으로 명세하고, Z/Eves 툴을 통해 메타모델의 정확성을 검사하는 것이다. 정형 명세는 클래스 모델과 Z/Object-Z간의 변환규칙을 적용해서 2+1 View 통합 메타모델의 구문과 정적 시멘틱에 대해 Z와 Object-Z 스키마로 각각 표현한다. 메타모델의 검사는 Z 스키마 명세에 대해 Z/Eves 도구를 사용하여 구문, 타입 검사 그리고 도메인 검사를 수행하여 메타모델의 정확성을 입증한다. 이로서, 2+1 View 통합 메타모델 메타모델의 Z/Object-Z 변환을 통해 구조물의 구문적 의미를 명확하게 표현할 수 있으며, 또한 그 메타모델의 정확성을 검사할 수 있다.

• 한국정보과학회논문지:컴퓨팅의 실제 및 레터
• /
• 제16권2호
• /
• pp.201-206
• /
• 2010

발전소와 임베디드 시스템, 지능형 빌딩과 같은 현대의 대부분의 시스템은 PLC 라는 특수목적 컴퓨터를 이용하여 자동제어된다. PLC 프로그래밍 언어 중에서 가장 많이 사용하는 것은 LD 프로그램이다. LD 프로그램의 검증은 시뮬레이션과 테스팅등을 통해 이루어지는데 이러한 방법은 프로그램 검증에 한계가 있다. 본 논문에는 LD 프로그램의 검증시 모델 체킹을 사용할 수 있도록 주어진 LD 프로그램을 모델 체커의 입력으로 자동으로 변환하는 방법을 기술한다. LD 프로그램과 SMV 모델의 의미구조를 정의 하고 이를 바탕으로 의미가 보존되는 변환 함수를 정형적으로 나타낸다.

• 디자인학연구
• /
• 제21권
• /
• pp.123-136
• /
• 1997

디자인은 예술적인 감성과 함께 과학적이며 객관적인 정보를 동시에 필요로 한다. 여러 학문을 통해 얻어진 정보가 조형화로의 번역에서는 디자이너에게 보다 풍부한 상상력을 요구한다. 본 연구에서는 특히 인간 공학 정보를 조형 문법과 요소로 옮기는 작업을 위해 종합적으로 연결시킬 수 있는 시스템의 필요성에 대한 인식과 함께 디자인 이미 론의 본 개념과 기능을 문헌을 통해 새롭게 조명하여 보았으며, 디자인 프로세스에서 인간공학의 의미와 문제점을 정리하였다. 그리고 이 두 학문이 하나의 시스템 안에서 효과적으로 조화되어 디자이너와 사용자에게 조형을 통한 만족을 줄 수 있는 이론적 모델로의 타당성을 탐색하는데 목적을 두었다.