• 한국정보과학회 언어공학연구회:학술대회논문집(한글 및 한국어 정보처리)
• /
• 한국정보과학회언어공학연구회 2020년도 제32회 한글 및 한국어 정보처리 학술대회
• /
• pp.529-534
• /
• 2020

인공신경망 기반 기계번역(Neural Machine Translation, NMT)이란 딥러닝(Deep learning)을 이용하여 출발 언어의 문장을 도착 언어 문장으로 번역해주는 시스템을 일컫는다. NMT는 종단간 학습(end-to-end learning)을 이용하여 기존 기계번역 방법론의 성능을 앞지르며 기계번역의 주요 방법론으로 자리잡게 됐다. 이러한 발전에도 불구하고 여전히 개체(entity), 또는 전문 용어(terminological expressions)의 번역은 미해결 과제로 남아있다. 개체나 전문 용어는 대부분 명사로 구성되는데 문장 내 명사는 주체, 객체 등의 역할을 하는 중요한 요소이므로 이들의 정확한 번역이 문장 전체의 번역 성능 향상으로 이어질 수 있다. 따라서 본 논문에서는 지식그래프(Knowledge Graph)를 이용하여 심볼릭 지식을 NMT와 결합한 뉴럴심볼릭 방법론을 제안한다. 또한 지식그래프를 활용하여 NMT의 성능을 높인 선행 연구 방법론을 한영 기계번역에 이용할 수 있도록 구조를 설계한다.

• 한국정보처리학회논문지
• /
• 제3권7호
• /
• pp.1781-1791
• /
• 1996

전문가시스템의 지식획득, 적합한 추론기구의 설계 및 구현, 지식의 정제 등 다단계 과정으로 이뤄져 있다. 각각을 하나의 연구이슈로 다양한 연구가 진행되어 왔으나 전체를 하나로 연계해 통합적 개발에 관해서는 상대적으로 연구가 활발히 진행되지 못한 실정이다. 지식획득은 전문가에 의해 수행되는 추론과정에서 특징 응용분야의 필요한 지식이 전달되어야 하므로 시식획득과 추론을 서로 밀접한 연관성을 갖는다. 지식의 정제는 추론과정에서 일어나는 문에의 제기와 이의 해결을 통해 지식베이스의 불완전하거나 논리적 모순을 찾아 해결함으로 지식베이스를 보다 완벽하고 정확한 것으로 만드는 것이다. ㅂㄴ 연구에서는 서로 연관된 다단계 과정이 통합적으로 개발될 수 있는 환경의 설저엥 대한 하나의 방안을 제시하려한다. 특히 도메인 모델이 잘 정립되기 어려운 분야에 학습기법을 활용햇 초기 지식 베이스를 구성할 수 있는 점진적 지식획득방법과 이를 통해 만들어진 지식베이스 규칙들을 학습기법의 일종인 개념적 클러스터링 기법을 이용하여 규칙모델을 구축하고 이를 이용해 효율적인 추론이 가능하게 하며, 지식획득 과정에서는 규칙의 오류를 제시할 수 있고 이에 대한 규칙의 수정이나 새로운 규칙이 기존의 지식구조에 합당한지를 결정하는 통합적 설계방안에 대해 연구한다. 지식의 정제는 설명기구와 규칙모델을 활용하여 문제의 원인을 찾고 해결점을 제시해 그에 대한 유효성을 검증합으로 이뤄지게 한다.뤄지게 한다.

• 한국펄프종이공학회:학술대회논문집
• /
• 한국펄프종이공학회 2002년도 춘계학술발표논문집
• /
• pp.76-84
• /
• 2002

Fracture toughness was considered as one of the good estimates of the paper break tendency of paper web in the press room. Paper break on the paper machine is caused by many factors such as paper machine irregular vibrations, impurities in the fiber furnish, shives, and so on. On the paper machine, the solid content of paper web is changing very rapidly from less than 1% to over 95%. We tried to measure the fracture toughness of paper web at different solid contents for providing the fundamental knowledge of paper break. Stretches of wet web were also measured and compared to the fracture toughness changes. Four different fiber furnishes (SwBKP, HwBKP, ONP, and OCC) were refined to different degrees, and at different solid contents (40%, 60%, 80%, and 95%), their fracture toughnesses were measured. Two fracture toughness measurement methods (essential work of fracture and Tryding's load-widening method) were used, and we found they gave identical results. The stretch curves of the wet webs against the axis of solid contents were very similar to the fracture toughness curves of those.

• 한국CDE학회논문집
• /
• 제5권2호
• /
• pp.127-135
• /
• 2000

Emergent computing paradigms such as genetic algorithms have found increased use in problems in engineering design. These computational tools have been shown to be applicable in the solution of generically difficult design optimization problems characterized by nonconvexities in the design space and the presence of discrete and integer design variables. Another aspect of these computational paradigms that have been lumped under the bread subject category of soft computing, is the domain of artificial intelligence, knowledge-based expert system, and machine learning. The paper explores a machine learning paradigm referred to as teaming classifier systems to construct the high-quality global function approximations between the design variables and a response function for subsequent use in design optimization. A classifier system is a machine teaming system which learns syntactically simple string rules, called classifiers for guiding the system's performance in an arbitrary environment. The capability of a learning classifier system facilitates the adaptive selection of the optimal number of training data according to the noise and multimodality in the design space of interest. The present study used the polynomial based response surface as global function approximation tools and showed its effectiveness in the improvement on the approximation performance.

• 한국농업기계학회:학술대회논문집
• /
• 한국농업기계학회 1996년도 International Conference on Agricultural Machinery Engineering Proceedings
• /
• pp.802-811
• /
• 1996

Farmers need alternatives for weed control due to the desire to reduce chemicals used in farming. However, conventional mechanical cultivation cannot selectively remove weeds located in the seedline between crop plants and there are no selective heribicides for some crop/weed situations. Since hand labor is costly , an automated weed control system could be feasible. A robotic weed control system can also reduce or eliminate the need for chemicals. Currently no such system exists for removing weeds located in the seedline between crop plants. The goal of this project is to build a real-time , machine vision weed control system that can detect crop and weed locations. remove weeds and thin crop plants. In order to accomplish this objective , a real-time robotic system was developed to identify and locate outdoor plants using machine vision technology, pattern recognition techniques, knowledge-based decision theory, and robotics. The prototype weed control system is composed f a real-time computer vision system, a uniform illumination device, and a precision chemical application system. The prototype system is mounted on the UC Davis Robotic Cultivator , which finds the center of the seedline of crop plants. Field tests showed that the robotic spraying system correctly targeted simulated weeds (metal coins of 2.54 cm diameter) with an average error of 0.78 cm and the standard deviation of 0.62cm.

• 펄프종이기술
• /
• 제33권5호
• /
• pp.37-44
• /
• 2001

Fracture toughness was considered as one of the good estimates of the paper break tendency of paper web in the press room. Paper break on the paper machine is caused by many factors such as paper machine irregular vibrations, impurities in the fiber furnish, shives, and so on. On the paper machine, the solid content of paper web is changing very rapidly from less than 1% to over 95%. We tried to measure the fracture toughness of paper web at different solid contents for providing the fundamental knowledge of paper break. Stretches of wet web were also measured and compared to the fracture toughness changes. Four different fiber furnishes (SwBKP, HwBKP, ONP, and OCC) were refined to different degrees, and at different solid contents (40%, 60%, 80% and 95%), their fracture toughnesses were measured. Two fracture toughness measurement methods (essential work of fracture and Tryding's load-widening method) were used, and we found they gave identical results. The stretch curves of the wet webs against the axis of solid contents were very similar to the fracture toughness curves of those.

• International Journal of Computer Science & Network Security
• /
• 제21권2호
• /
• pp.229-237
• /
• 2021

Our modern 'information-hungry' age demands delivery of information at unprecedented fast rates. Timely delivery of noteworthy information about recent events can help people from different segments of life in number of ways. As world has become global village, the flow of news in terms of volume and speed demands involvement of machines to help humans to handle the enormous data. News are presented to public in forms of video, audio, image and text. News text available on internet is a source of knowledge for billions of internet users. Urdu language is spoken and understood by millions of people from Indian subcontinent. Availability of online Urdu news enable this branch of humanity to improve their understandings of the world and make their decisions. This paper uses available online Urdu news data to train machines to automatically categorize provided news. Various machine learning algorithms were used on news headline for training purpose and the results demonstrate that Bernoulli Naïve Bayes (Bernoulli NB) and Multinomial Naïve Bayes (Multinomial NB) algorithm outperformed other algorithms in terms of all performance parameters. The maximum level of accuracy achieved for the dataset was 94.278% by multinomial NB classifier followed by Bernoulli NB classifier with accuracy of 94.274% when Urdu stop words were removed from dataset. The results suggest that short text of headlines of news can be used as an input for text categorization process.

• Journal of Nutrition and Health
• /
• 제55권1호
• /
• pp.1-9
• /
• 2022

In the past few decades, great progress has been made on understanding the interaction between nutrition and health status. But despite this wealth of knowledge, health problems related to nutrition continue to increase. This leads us to postulate that the continuing trend may result from a lack of consideration for intra-individual biological variation on dietary responses. Precision nutrition utilizes personal information such as age, gender, lifestyle, diet intake, environmental exposure, genetic variants, microbiome, and epigenetics to provide better dietary advices and interventions. Recent technological advances in the artificial intelligence, big data analytics, cloud computing, and machine learning, have made it possible to process data on a scale and in ways that were previously impossible. A big data platform is built by collecting numerous parameters such as meal features, medical metadata, lifestyle variation, genome diversity and microbiome composition. Sophisticated techniques based on machine learning algorithm can be used to integrate and interpret multiple factors and provide dietary guidance at a personalized or stratified level. The development of a suitable machine learning algorithm would make it possible to suggest a personalized diet or functional food based on analysis of intra-individual metabolic variation. This novel precision nutrition might become one of the most exciting and promising approaches of improving health conditions, especially in the context of non-communicable disease prevention.

• International Journal of Computer Science & Network Security
• /
• 제24권3호
• /
• pp.113-124
• /
• 2024

One of the goals of the Saudi Arabia 2030 vision is to ensure full employment of its citizens. Recruitment of graduates depends on the quality of skills that they may have gained during their study. Hence, the quality of education and ensuring that graduates have sufficient knowledge about the in-demand skills of the market are necessary. However, IT graduates are usually not aware of whether they are suitable for recruitment or not. This study builds a prediction model that can be deployed on the web, where users can input variables to generate predictions. Furthermore, it provides data-driven recommendations of the in-demand skills in the Saudi IT labor market to overcome the unemployment problem. Data were collected from two online job portals: LinkedIn and Bayt.com. Three machine learning algorithms, namely, Support Vector Machine, k-Nearest Neighbor, and Naïve Bayes were used to build the model. Furthermore, descriptive and data analysis methods were employed herein to evaluate the existing gap. Results showed that there existed a gap between labor market employers' expectations of Saudi workers and the skills that the workers were equipped with from their educational institutions. Planned collaboration between industry and education providers is required to narrow down this gap.

• 한국프랜차이즈경영연구
• /
• 제12권4호
• /
• pp.1-11
• /
• 2021

Purpose: The franchise system started by Singer Sewing Machine in the US is acting as a national economic growth engine in terms of job creation and economic growth. In China, the franchise system was introduced in the mid-1980s. And since joining the WTO, it has grown by 5-6% every year. However, compared to the growth rate of franchises, studies on shared growth between the chain headquarters and franchisees were insufficient. Accordingly, recent studies related to shared growth between the chain headquarters and franchisees have been active in China. The purpose of this study is to examine the knowledge transfer system between the knowledge creation, knowledge sharing, and the use of knowledge by franchise chain headquarters in China. In addition, the relationship between franchise satisfaction and performance is identified. Research design, data, and methodology: The data were collected from franchise stores in Sichuan, China, and were conducted with the help of ○○ Incubation, a Sichuan Province-certified incubator. From November 2020 to January 2021, 350 copies of the questionnaire were distributed in China, and 264 copies were returned. Of these, 44 copies with insincere answers and response errors were excluded, and 222 copies were used for analysis. The data were analyzed with SPSS 22.0 and AMOS 22.0 statistical packages. Result: The results of this study are as follows. First, knowledge creation has been shown to have a statistically significant impact on knowledge sharing and knowledge utilization. In particular, the effectiveness of knowledge creation was higher in knowledge sharing than in knowledge utilization. And we can see that knowledge sharing also has a statistically significant e ffect on knowledge utilization. Second, knowledge sharing was not significant for transaction satisfaction and business performance, and knowledge utilization was significant for transaction satisfaction and business performance. These results can be said to mean less interdependence of the Chinese franchise system. Finally, transaction satisfaction was statistically significant to business performance. The purpose of this study was to examine the importance of knowledge management to secure long-term competitive advantage for Chinese franchises. This study shows that knowledge sharing is important for long-term franchise growth. And we can see that there is a lack of knowledge sharing methods in the case of franchises in China. I n addition, it was found that the growth of Chinese franchises requires systematization of communication, information sharing measures and timing, help from chain headquarters, and mutual responsibility awareness.