• Annual Conference on Human and Language Technology
• /
• 2010.10a
• /
• pp.3-8
• /
• 2010

본 연구에서는 일련의 연구에서 수집된 영작문 오류 유형의 정제된 자료를 토대로 연관 규칙을 생성하고, 학습을 통해서 효용성이 검증된 연관 규칙을 활용해서 영작문 데이터의 형태 통사 오류를 자동으로 탐지한다. 영작문 데이터에서 형태 통사 오류를 찾아내는 작업은 많은 시간과 자원이 소요되는 작업이므로 자동화가 필수적이다. 기존의 연구들이 통계적 모델을 활용한 어휘적 오류에 치중하거나 언어 이론적 틀에 근거한 통사 처리에 집중하는 반면에, 본 연구는 데이터 마이닝을 통해서 정제된 데이터에서 연관 규칙을 생성하고 이를 검증한 후 형태 통사 오류를 감지한다. 이전 연구들에서는 이론적 틀에 맞추어진 규칙 생성이나 언어 모델 생성을 위한 대량의 코퍼스 데이터와 같은 다량의 지식 베이스 생성이 필수적인데, 본 연구는 적은 양의 정제된 데이터를 활용한다. 영작문 오류 유형의 형태 통사 연관 규칙을 생성하기 위해서 Apriori 알고리즘을 활용하였다. 알고리즘을 통해서 생성된 연관 규칙 중 잘못된 규칙이 생성될 가능성이 있으므로, 상관성 검정, 코사인 유사도와 같은 규칙 효용성의 통계적 검증을 활용해서 타당한 규칙만을 학습하였다. 이를 통해서 축적된 연관 규칙들을 영작문 오류를 자동으로 탐지하는 실험에 활용하였다.

• Annual Conference on Human and Language Technology
• /
• 1995.10a
• /
• pp.43-49
• /
• 1995

대부분 자연 언어 이해 시스템이나 중간 언어 방식을 이용한 기계 번역 시스템에서 자연 언어 문장을 이해하고 번역하기 위해서는 대량의 지식을 이용한 의미 구조의 생성이 요구된다. 따라서 개념 그래프를 이용하여 한국어 문장의 내부 의미 구조를 생성하기 위해서는 각 단어에 해당하는 개념과 개념들 사이의 개념적 관계를 나타내는 지식들이 요구된다. 그러나 이를 위한 의미 구조 생성 방법과 요구되는 지식 베이스를 정확하게 구축하는 것은 어렵고 응용 도메인에 종속한다는 문제가 있다. 본 논문에서 우리는 문형 이론을 도입하여 문형을 중심으로 의미 구조 생성을 위한 변환 규칙을 설계하고 이를 이용하여 방대한 지식 베이스의 구축없이 의미 구조를 생성할 수 있는 방법에 논의한다. 또한 본 논문에서 문형을 이용하면 몇 가지 모호성 문제를 해결할 수 있음을 보이고 문형의 한계에 대해서도 살펴본다.

• Korean Journal of Cognitive Science
• /
• v.16 no.4
• /
• pp.369-386
• /
• 2005

Although Intelligent Tutoring System(ITS) offers individualized learning environment that overcome limited function of existent CAI, and consider many learners' variable, there is little development to be using at the sites of schools because of inefficiency of investment and absence of pedagogical content knowledge representation techniques. To solve these problem, we should study a method, which represents knowledge for ITS, and which reuses knowledge base. On the pedagogical content knowledge, the knowledge in education differs from knowledge in a general sense. In this paper, we shall primarily address the multi-complex structure of knowledge and explanation of learning vein using multi-complex structure. Multi-Complex, which is organized into nodes, clusters and uses by knowledge base. In addition, it grows a adaptive knowledge base by self-learning. Therefore, in this paper, we propose the 'Extended Neural Logic Network(X-Neuronet)', which is based on Neural Logic Network with logical inference and topological inflexibility in cognition structure, and includes pedagogical content knowledge and object-oriented conception, verify validity. X-Neuronet defines that a knowledge is directive combination with inertia and weights, and offers basic conceptions for expression, logic operator for operation and processing, node value and connection weight, propagation rule, learning algorithm.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2001.10a
• /
• pp.233-236
• /
• 2001

멀티미디어 데이터의 증가와 마이닝 기술의 발전으로 인해 멀티미디어 마이닝에 대한 관심이 증가하고 있다. 본 논문에서는 특성국지화를 이용한 내용기반의 정보검색 기술과 다차원 데이터큐브 구축기술을 통해 멀티미디어 데이터에서 연관규칙을 찾아내는 멀티미디어 데이터마이닝 시스템 프로토타입을 제안한다. 특히 멀티미디어 데이터의 칼라, 질감 등 거시적인 이미지 성분 대신 이미지의 영역성과 유사성을 이용한 특성국지화방법을 이용하여 이미지를 분할함으로써 방대한 데이타에서 효과적인 내용기반의 정의 검색을 시행하고 검색한 벡터를 메타데이타로 한 데이스베이스를 구축한다. 그리고 데이터베이스에서 데이터간 연관규칙을 찾아내어 지식을 마이닝하는데 효과적인 다차원 데이터큐브를 구축하고 여기에 연관규칙 검색 알고리즘을 적용한다.

• Journal of Intelligence and Information Systems
• /
• v.13 no.3
• /
• pp.17-29
• /
• 2007

Inference engine that performs the brain of software agent in next generation's web with various standards based on standard language of the web, XML has to understand SWRL (Semantic Web Rule Language) that is a language to express the rule in the Semantic Web. In this research, we want to develop a forward inference engine, SMART-F (SeMantic web Agent Reasoning Tools-Forward chaining inference engine) that uses SWRL as a rule express method, and OWL as a fact express method. In the traditional inference field, the Rete algorithm that improves effectiveness of forward rule inference by converting if-then rules to network structure is often used for forward inference. To apply this to the Semantic Web, we analyze the required functions for the SWRL-based forward inference, and design the forward inference algorithm that reflects required functions of next generation's Semantic Web deducted by Rete algorithm. And then, to secure each platform's independence and portability in the ubiquitous environment and overcome the gap of performance, we developed management tool of fact and rule base and forward inference engine. This is compatible with fact and rule base of SMART-B that was developed. So, this maximizes a practical use of knowledge in the next generation's Web environment.

• Journal of KIISE:Software and Applications
• /
• v.37 no.4
• /
• pp.282-296
• /
• 2010

OWL is an ontology language for the Semantic Web, and suited to modelling the knowledge of a specific domain in the real-world. Ontology also can infer new implicit knowledge from the explicit knowledge. However, the modeled knowledge cannot be complete as the whole of the common-sense of the human cannot be represented totally. Ontology do not concern handling nonmonotonic reasoning to detect incomplete modeling such as the integrity constraints and exceptions. A default rule can handle the exception about a specific class in ontology. Integrity constraint can be clear that restrictions on class define which and how many relationships the instances of that class must hold. In this paper, we propose a practical reasoning system for open and closed-world reasoning that supports a novel hybrid integration of ontology based on open world assumption (OWA) and non-monotonic rule based on closed-world assumption (CWA). The system utilizes a method to solve the problem which occurs when dealing with the incomplete knowledge under the OWA. The method uses the answer set programming (ASP) to find a solution. ASP is a logic-program, which can be seen as the computational embodiment of non-monotonic reasoning, and enables a query based on CWA to knowledge base (KB) of description logic. Our system not only finds practical cases from examples by the Protege, which require non-monotonic reasoning, but also estimates novel reasoning results for the cases based on KB which realizes a transparent integration of rules and ontologies supported by some well-known projects.

• The Transactions of the Korea Information Processing Society
• /
• v.4 no.12
• /
• pp.3010-3022
• /
• 1997

Expert system is one of AI area which was came out at the end of 19705s. It simulates the human's way of thinking to give solutions of Problem in many applications. Most expert system consists of many components such as inference engine, knowledge base, and so on. Especially the performance of expert system depends on the control of enfficiency of inference engine. Inference engine has to get features; tirst, if possible to minimize restrictions when the knowledge base is constructed second, it has to serve various kinds of inferencing methods. In this paper, we design and implement the inference engine which is able to support the general functions to knowledge domain and inferencing method. For the purpose, forward chaining, backward chaining, and direct chaining was employed as an inferencing method in order to be able to be used by user request selectively. Also we not on1y selected production system which makes one ease staradization and modulation to obtain knowledges in target domain, but also constructed knowledge base by means of Extended Clause Bit Metrics (ECBM). Finally, the performance evaluation of inference engine between Rete pattern matching and ECBM has been done.

• Journal of the Korea Convergence Society
• /
• v.9 no.10
• /
• pp.75-80
• /
• 2018

In this paper, we propose heterogeneous lifelog mining model in health big-data platform. It is an ontology-based mining model for collecting user's lifelog in real-time and providing healthcare services. The proposed method distributes heterogeneous lifelog data and processes it in real time in a cloud computing environment. The knowledge base is reconstructed by an upper ontology method suitable for the environment constructed based on the heterogeneous ontology. The restructured knowledge base generates inference rules using Jena 4.0 inference engines, and provides real-time healthcare services by rule-based inference methods. Lifelog mining constructs an analysis of hidden relationships and a predictive model for time-series bio-signal. This enables real-time healthcare services that realize preventive health services to detect changes in the users' bio-signal by exploring negative or positive correlations that are not included in the relationships or inference rules. The performance evaluation shows that the proposed heterogeneous lifelog mining model method is superior to other models with an accuracy of 0.734, a precision of 0.752.

• Journal of KIISE:Databases
• /
• v.28 no.3
• /
• pp.301-314
• /
• 2001

Data mining refers to a set of techniques for discovering implicit and useful knowledge from large database. Many studies on data mining have been pursued and some of them have involved issues of temporal data mining for discovering knowledge from temporal database, such as sequential pattern, similar time sequence, cyclic and temporal association rules, etc. However, all of the works treat problems for discovering temporal pattern from data which are stamped with time points and do not consider problems for discovering knowledge from temporal interval data. For example, there are many examples of temporal interval data that it can discover useful knowledge from. These include patient histories, purchaser histories, web log, and so on. Allen introduces relationships between intervals and operators for reasoning about relations between intervals. We present a new data mining technique that can discover temporal relation rules in temporal interval data by using the Allen's theory. In this paper, we present two new algorithms for discovering algorithm for generating temporal relation rules, discovers rules from temporal interval data. This technique can discover more useful knowledge in compared with conventional data mining techniques.

• Journal of KIISE:Software and Applications
• /
• v.28 no.11
• /
• pp.795-813
• /
• 2001

Sophisticated geometric structure analysis must be preceded to create electronic document from logical components extracted from document image. this paper presents a knowledge-based method for sophisticated geometric structure analysis of technical journal pages. The proposed knowledge base encodes geometric characteristics that are not only common in technical journals but also publication-specific in the form rules. The method takes the hybrid of top-down and bottom-up techniques and consists of two phases: region segmentation and identification. Generally, the result of segmentation process does not have a one-to-one matching with composite layout components. Therefore, the proposed method identifies non-text objects such as image, drawing and table, as well as text objects such as text line and equation by splitting or grouping segmented regions into composite layout components. Experimental results with 372 images scanned from the IEEE Transactions on Pattern Analysis and Machine Intelligence show that the proposed method has performed geometrical structure analysis successfully on more than 99% of the test images, resulting in sophisticated performance compared with previous works.