• Annual Conference on Human and Language Technology
• /
• 2015.10a
• /
• pp.194-197
• /
• 2015

Long Short-term Memory Network(LSTM) 기반 Recurrent Neural Network(RNN)는 순차 데이터를 모델링 할 수 있는 딥 러닝 모델이다. 기존 RNN의 그래디언트 소멸 문제(vanishing gradient problem)를 해결한 LSTM RNN은 멀리 떨어져 있는 이전의 입력 정보를 볼 수 있다는 장점이 있어 음성 인식 및 필기체 인식 등의 분야에서 좋은 성능을 보이고 있다. 또한 LSTM RNN 모델에 의존성(전이 확률)을 추가한 LSTM CRF모델이 자연어처리의 한 분야인 개체명 인식에서 우수한 성능을 보이고 있다. 본 논문에서는 한국어 문장의 지배소가 문장 후위에 나타나는 점에 착안하여 Backward 방식의 LSTM CRF 모델을 제안하고 이를 한국어 의미역 결정에 적용하여 기존 연구보다 더 높은 성능을 얻을 수 있음을 보인다.

• Proceedings of the Korean Society for Agricultural Machinery Conference
• /
• 1993.10a
• /
• pp.1147-1156
• /
• 1993

In order to analyze lateral control in the backward maneuver of a tractor -trailer combination , a kinematic vehicle model and a human operator model with time delay were utilized for the operator/vehicle system. The analysis was carried out using the frequency domain approach. The open-loop stability of the vehicle motion was analyzed through the transfer functions. The sensitivity of the stability of the vehicle motion. to a change in the steering angle, was also analyzed. A mathematical model of the closed -loop operator/vehicle system was then formulated. The closed -loop stability of the operator /vehicle system was then analyzed. The effect of the delay time on the system was also analyzed through computer simulation.

• Journal of Internet Computing and Services
• /
• v.5 no.4
• /
• pp.115-124
• /
• 2004

This paper presents a weighted fuzzy backward reasoning algorithm for rule-based systems based on weighted fuzzy Petri nets. The fuzzy production rules in the knowledge base of a rule-based system are modeled by weighted fuzzy Petri nets, where the truth values of the propositions appearing in the fuzzy production rules and the certainty factors of the rules are represented by fuzzy numbers. Furthermore, the weights of the propositions appearing in the rules are also represented by fuzzy numbers. The proposed weighted fuzzy backward reasoning generates the backward reasoning path from the goal node to the initial nodes and then evaluates the certainty factor of the goal node. The algorithm we proposed can allow the rule-based systems to perform weighted fuzzy backward reasoning in more flexible and human-like manner.

• Human Ecology Research
• /
• v.61 no.3
• /
• pp.297-318
• /
• 2023

The purpose of this study was to develop a high school home economics financial consumer education program based on backward design and validation by experts. The program was designed and developed by first selecting learning content elements through a review of existing research and an analysis of relevant literature. The next step was to categorize these elements into seven themes and apply the backward design instructional design model 2.0. The program was prepared in the form of a 21st teaching-learning course plan and workbook and was verified by nine home economics teachers with working experience in high school. The evaluation revealed that the average value for all questions was 3.81 (out of 4 points) and the CVR was .99, indicating that the program was valid. In addition, positive evaluations were received in terms of learning goals, content level, and learner participation by class. This study has significance in that a systematic financial consumer education program was developed by Education of Home Economics to improve the financial literacy of high school students. It can therefore be used as an elective course (mini-course) in Home Economics in the high school credit system. A follow-up study will be required to assess the improvement in financial literacy after implementing this program.

• Journal of Intelligence and Information Systems
• /
• v.12 no.2
• /
• pp.67-81
• /
• 2006

While the existing Web focuses on the interface with human users based on HTML, the next generation Web will focus on the interaction among software agents by using XML and XML-based standards and technologies. The inference engine, which will serve as brains of software agents in the next generation Web, should thoroughly understand the Semantic Web, the standard language of the next generation Web. As abasis for the service, the W3C (World Wide Web Consortium) has recommended SWRL (Semantic Web Rule Language) which had been made by compounding OWL (Web Ontology Language) and RuleML (Rule Markup Language). In this research, we develop a backward chaining inference engine SMART-B (SeMantic web Agent Reasoning Tools -Backward chaining inference engine), which uses SWRL and OWL to represent rules and facts respectively. We analyze the requirements for the SWRL-based backward chaining inference and design analgorithm for the backward chaining inference which reflects the traditional backward chaining inference algorithm and the requirements of the next generation Semantic Web. We also implement the backward chaining inference engine and the administrative tools for fact and rule bases into Java components to insure the independence and portability among different platforms under the environment of Ubiquitous Computing.

• Journal of Korea Multimedia Society
• /
• v.7 no.4
• /
• pp.559-566
• /
• 2004

In general, the certainty factors of the fuzzy production rules and the certainty factors of fuzzy propositions appearing in the rules are represented by real values between zero and one. If it can allow the certainty factors of the fuzzy production rules and the certainty factors of fuzzy propositions to be represented by interval -valued fuzzy sets, then it can allow the reasoning of rule-based systems to perform fuzzy reasoning in more flexible manner. This paper presents fuzzy Petri nets and proposes an interval-valued fuzzy backward reasoning algorithm for rule-based systems based on fuzzy Petri nets Fuzzy Petri nets model the fuzzy production rules in the knowledge base of a rule-based system, where the certainty factors of the fuzzy propositions appearing in the fuzzy production rules and the certainty factors of the rules are represented by interval-valued fuzzy sets. The algorithm we proposed generates the backward reasoning path from the goal node to the initial nodes and then evaluates the certainty factor of the goal node. The proposed interval-valued fuzzy backward reasoning algorithm can allow the rule-based systems to perform fuzzy backward reasoning in a more flexible and human-like manner.

• Journal of the Ergonomics Society of Korea
• /
• v.33 no.2
• /
• pp.155-165
• /
• 2014

Objective: The aim of this study is to develop an office chair enabling to keep working at reclined sitting posture. Background: Sedentary workers are supposed to change the posture frequently during long hours of sitting. A reclined sitting position has been recommended to reduce disc pressure. But slumped sitting posture caused by the buttock sliding forward without any adjustment of back reclining is commonly observed. The worker seems to have tendency to change the sitting posture maintaining working condition. We assumed the reason to be their hands movement away from the working space when tilting backward. Method: Slide mechanism allowing seat to move forward was designed to maintain the hand position in working space during reclining. A prototype was manufactured and tilting motion was analyzed using motion capture system. Four experiment chairs were tested including the manufactured prototype chair and three other commercial chairs. Results: A backward movements of the hand position were 13.0mm, 101.7mm, 156.1mm and 139.3mm at the prototype chair, compared to chair B, chair C and chair D, respectively. And the movement was remarkably small at the prototype chair. Conclusion: The developed seat sliding chair allows back tilting maintaining hand position at working space. We expect the user tilting back more often than normal tilting chair during seated work. But further investigation is required to figure out the effectiveness of the developed chair using prolonged working hours. Application: The developed office chair directly affects commercialization.

• 제어로봇시스템학회:학술대회논문집
• /
• 2001.10a
• /
• pp.59.4-59
• /
• 2001

In this paper, wireless interface of the motion between human and robot is implemented. The idea is that if a human who is equiped with device including accelerometer and rate-gyro sensor move his/her arm, then the robot follows human motion. The robot is designed as wheeled type mobile robot with two link arms. The robot´s basic movements such as forward, backward, left, right movement can be controlled from foot sensor which human steps on. Arm movements can be controlled by arm motion of human motion. In order to detect human motion, sensor data analysis from gyro and accelerometer has to be done. Data from sensors are transferred through wireless communication to activate the robot.

• The Transactions of the Korea Information Processing Society
• /
• v.5 no.4
• /
• pp.951-958
• /
• 1998

In this paper, we propose a backward reasoning algorithm which can be utilized in the fuzzy Petri-net representation representing fuzzy production rules. The fuzzy Petri-net representation can be used to model a approximate reasoning system and implement a fuzzy inference engine. The proposed algorithm, which uses the proper belief evaluation functions according to fuzzy concepts in antecedentes and consequents of fuzzy production rules, is more closer to human intuition and reasoning than other methods. This algorithm generates the backward reasoning path from the goal to the initial nodes and evaluates the belief value of the goal node using belief evaluation functions.

• Korean Journal of Cognitive Science
• /
• v.10 no.4
• /
• pp.51-62
• /
• 1999

This paper proposes a fuzzy Petri net representation to represent the fuzzy production rules of a rule-based expert system. Based on the fuzzy Petri net representation. we present a fuzzy reasoning algorithms which consist of forward and b backward reasoning algorithm. The proposed algorithms. which use the proper belief evaluation functions according to fuzzy concepts in antecedent and consequent of a fuzzy production rule. are more closer to human intuition and reasoning than other methods. The forward reasoning algorithm can be represented by a reachability tree as a kind of finite directed tree. The backward reasoning algorithm generates the backward reasoning path from the goal to the initial nodes and then evaluates the belief value of the goal node using belief evaluation functions.