• Journal of Information Science Theory and Practice
• /
• 제1권1호
• /
• pp.10-26
• /
• 2013

Domain transfer is a widely recognized problem for machine learning algorithms because models built upon one data domain generally do not perform well in another data domain. This is especially a challenge for tasks such as opinion classification, which often has to deal with insufficient quantities of labeled data. This study investigates the feasibility of self-training in dealing with the domain transfer problem in opinion classification via leveraging labeled data in non-target data domain(s) and unlabeled data in the target-domain. Specifically, self-training is evaluated for effectiveness in sparse data situations and feasibility for domain adaptation in opinion classification. Three types of Web content are tested: edited news articles, semi-structured movie reviews, and the informal and unstructured content of the blogosphere. Findings of this study suggest that, when there are limited labeled data, self-training is a promising approach for opinion classification, although the contributions vary across data domains. Significant improvement was demonstrated for the most challenging data domain-the blogosphere-when a domain transfer-based self-training strategy was implemented.

• ETRI Journal
• /
• 제36권3호
• /
• pp.429-438
• /
• 2014

Semantic role labeling (SRL) is a task in natural-language processing with the aim of detecting predicates in the text, choosing their correct senses, identifying their associated arguments, and predicting the semantic roles of the arguments. Developing a high-performance SRL system for a domain requires manually annotated training data of large size in the same domain. However, such SRL training data of sufficient size is available only for a few domains. Constructing SRL training data for a new domain is very expensive. Therefore, domain adaptation in SRL can be regarded as an important problem. In this paper, we show that domain adaptation for SRL systems can achieve state-of-the-art performance when based on structural learning and exploiting a prior model approach. We provide experimental results with three different target domains showing that our method is effective even if training data of small size is available for the target domains. According to experimentations, our proposed method outperforms those of other research works by about 2% to 5% in F-score.

• 과학교육연구지
• /
• 제41권3호
• /
• pp.462-479
• /
• 2017

본 연구는 중학교 과학교과서 2종에 수록된 시각화자료들을 데이터시각화와 인포그래픽으로 나누어 분류한 후, 인포그래픽 분석틀에 따라 에너지, 물질, 생명, 지구의 과학 내용 영역별로 인포그래픽의 유형과 특징을 분석하였다. 연구 결과는 다음과 같다. 첫째, 중학교 과학 교과서에서 시각화 자료는 인포그래픽(평균 23.20%)보다 데이터시각화(평균 76.80%)에 해당하는 자료가 더 많았다. 둘째, 인포그래픽은 정보내용의 영역에서 과학 교과의 영역별로 서로 다른 특징을 보였다. 에너지와 물질 영역에서는 정보내용 중 '관계'에 해당하는 내용이 많았고, 생명 영역에서는 '기능', 지구 영역에서는 '위치'와 관련된 정보내용이 많이 나타났다. 과학 전체 영역에서 시각표현 유형은 일러스트형과 비교분석형이 주로 사용되었으며 레이아웃 유형은 수평형이 가장 많은 비율을 차지하였다. 따라서 과학교과서에서 제시하는 그림은 단순 정보를 전달하는 데이터시각화보다 인포그래픽으로 표현하여 제시할 필요가 있으며, 과학의 각 영역별로 다양한 인포그래픽 유형을 활용한 교과서 개발이 필요하다.

• Smart Structures and Systems
• /
• 제8권3호
• /
• pp.321-341
• /
• 2011

Acoustic emission (AE) monitoring was conducted for mortar specimens under three types of static loading patterns (cubic-splitting, direct-shear and pull-out). Each of the applied loading patterns was expected to produce a particular fracture process. Subsequently, the AEs generated by various fracture or damage processes carried specific information on temporal micro-crack behaviors of concrete for post analysis, which was represented in the form of detected AE signal characteristics. Among various available characteristics of acquired AE signals, frequency content was of great interest. In this study, cement-based piezoelectric sensor (as AE transducer) and home-programmed DEcLIN monitoring system were utilized for AE monitoring on mortar. The cement-based piezoelectric sensor demonstrated enhanced sensitivity and broad frequency domain response range after being embedded into mortar specimens. This broad band characteristic of cement-based piezoelectric sensor in frequency domain response benefited the analysis of frequency content of AE. Various evaluation methods were introduced and employed to clarify the variation characteristics of AE frequency content in each test. It was found that the variation behaviors of AE frequency content exhibited a close relationship with the applied loading processes during the tests.

• Kyungpook Mathematical Journal
• /
• 제45권1호
• /
• pp.131-135
• /
• 2005

Let an integral domain R be locally polynomial over an integral domain D and let R be a content module over D. We show that if D is a PVMD, then $$Cl_t(R){\sim_=}Cl_t(D)$$. This generalizes the polynomial case. We also show that R is a PVMD if and only if D is a PVMD if and only if $R_{N_v}$ is a PVMD.

• 한국초등과학교육학회지:초등과학교육
• /
• 제25권2호
• /
• pp.149-158
• /
• 2006

The purpose of this study was to analyze the analogies used in elementary science textbooks. For this, the analogies were classified into the following criteria : analogy-related contents included in elementary science text-books, types of representation style, and the role of analogy. The results of this study were as follows : the total of analogies was 154 in the 6th and 166 in the 7th curriculum. Most of them were expressed in textbooks and experimental texts. More structural/functional, verbal/pictorial, concrete/concrete, concrete/abstract, simple and enriched analogies were included in the science textbooks of the 7th than those of the 6th. Most of them took the role as an aid to understanding. According to the domain of science content, energy, material, life domains were increased in the 7th than the 6th, with the exception of the earth science domain. The ratio of used analogies in the life domain was higher than the others, and in the material domain it was the lowest. From these results, it can be said that, in order to help students' understanding, further research into the application of analogies is needed.

• 기본간호학회지
• /
• 제9권2호
• /
• pp.180-189
• /
• 2002

Purpose : This study focused on affirming of the effects of Dan-jun breathing exercise experienced by women in midlife. Method: The data were collected via direct interview from 20 participants, who participated in a Dan-Jun exercise program during three months. Collected data were analyzed by content analysis. Result : 1. As for the health problems of women in midlife before the Dan-jun exercise program, 93 items were extracted in the content analysis These items were categorized into 59 attributes and 15 higher attributes. 1) Of the 59 attributes of health problems, shoulder pain (30%) was the most dominant. 2) The 15 higher attributes consisted for the physical domain of, weakness, pain, joint stiffness, sexual problem, sleep disorder, gastro-intestinal disorder. menstrual irregularity. circulatory disorder, respiratory disorder. constipation, urinary incontinence, and for the psychological domain, anger, emptiness, depression, and anxiety. 2. As for the effects experienced by women in midlife after the Dan-jun exercise program, 169 items were extracted in the content analysis. These items were categorized into 85 attributes and 14 higher attributes. 1) Of 85 attributes, lightening of physical condition (55%) was the most dominant. 2) The higher attributes consisted in physical domain of, recovery of vigor, pain relief, menstrual regularity and improvements in flexibility. sexual problem, sleep disorder, gastro-intestinal disorder, circulatory disorder constipation, respiratory disorder and urinary incontinence, and for the psychological domain, relaxation, cultured mind and self-confidence. Conclusions : Dan-jun breathing exercise program for three months showed positive effects on physical and psychological health in women in midlife. In this context, it can be also considered as a significant nursing intervention for maintenance and promotion of the health of these women.

• 한국초등과학교육학회지:초등과학교육
• /
• 제33권3호
• /
• pp.491-509
• /
• 2014

The purpose of this study was to compare and analyze the elementary science curriculum of Finland, which ranked at the first place in the science domain of Programme for International Student Assessment (PISA), a comparative study on the academic achievements of OECD member nations, for three consecutive years and recorded a high national competitiveness index, and that of South Korea, thus providing information needed to develop curriculums and textbooks in the middle of occasional curriculum revisions and giving useful implications for the implementation of curriculum in the field of education in South Korea. The research findings were as follows: First, as for the science content areas based on the ratio of large area items according to the evaluation and analysis framework of TIMSS 2007, South Korea's ratio of life, chemistry, physics, and earth hardly showed fitness for TIMSS 2007 and exhibited equal distribution among the areas. In Finland, the ratio of life, chemistry, physical, and earth was similar to the fourth grade level of TIMSS 2007. The country showed differential distribution with life accounting for the highest percentage. Second, as for the cognitive domains, South Korea showed a high percentage in "Uses and Procedures of Tools" of "1. Knowing" and "Making Connections," "Comparison/Contrast/Classification," and "Uses of Models" of "2. Application." Finland recorded a high percentage in "Information Interpretation" of "2. Application." While South Korea focused on the uses and methods of scientific instruments during scientific activities, Finland made an approach with a focus on problems related to daily life such as the interpretation of information including reports and graphs from an interpretative perspective.

• 한국컴퓨터정보학회논문지
• /
• 제5권3호
• /
• pp.215-219
• /
• 2000

The purpose of this paper is to propose a method to maximize the efficiency of a content-based image retrieval for various kinds of images. This paper discuss the self-adaptivity for the change of image domain and the self-selection of optimal features for query image, and present the efficient method to maximize content-based retrieval for various kinds of images. In this method, a content-based retrieval system is adopted to select automatically distinctive feature patterns which have a maximum efficiency of image retrieval in various kinds of images. Experimental results show that the Proposed method is improved 3% than the method using individual features.

• 한국초등과학교육학회지:초등과학교육
• /
• 제35권2호
• /
• pp.166-180
• /
• 2016

This paper examines the effects on academic achievement, scientific process skills and affective domain for elementary students learning the 'Chicken life cycle' through traditional science class versus a smart media based STEAM approach. Students designed and built a hatching jar and created a smart media content for chickens using time-lapse technology. This STEAM program was developed to improve their scientific concepts of animals over nine periods of classes using integrated education methods. The experimental study took place in the third grade of public schools in a province, with the STEAM approach applied in 2 classes (44 students) and the traditional discipline approach implemented in 2 classes (46 students). The STEAM education significantly influenced the improvement of academic achievements, basic scientific process skills and affective domain. The results suggest that this STEAM approach for teaching scientific concepts of animal life cycles has the performance in terms of knowledge, skills and affect gain achievements in elementary school students' learning when compared to a traditional approach. Moreover, the smart media based STEAM program is helpful to lead students to engage in integrated problem-solving designs and learning science and technology.