• The Mathematical Education
• /
• v.54 no.3
• /
• pp.261-282
• /
• 2015

The purpose of this study was to provide insights on making Korean mathematics framework by analytical comparison of three major assessments such as the NAEP 2015, the TIMSS 2015 and the PISA 2015. This study focused on the key differences and common themes of mathematics frameworks among three major assessments. In order to achieve this purpose, mathematical frameworks of the NAEP 2015, the TIMSS 2015, and the PISA 2015 were analyzed and compared. The criteria of the comparison were content domain and cognitive domain. The comparing criteria of content domain were based on NCTM content standards and cognitive domain were used the three understanding levels of Jan de Lange's pyramid model. Based on these comparisons, researchers discussed that Korea mathematical framework was needed to have a set of content categories that reflect the range of underlying mathematical phenomena and a set of cognitive levels which contain the range of underlying fundamental mathematical capabilities including consideration of contexts.

• ETRI Journal
• /
• v.36 no.3
• /
• 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.

• Journal of Information Science Theory and Practice
• /
• v.1 no.1
• /
• 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
• /
• v.30 no.3
• /
• pp.394-402
• /
• 2008

Domain analysis is the process of analyzing related software systems in a domain to find their common and variable parts. In the case of device drivers, they are highly suitable for domain analysis because device drivers of the same domain are implemented similarly for each device and each system that they support. Considering this characteristic, this paper introduces a new approach to the domain analysis of device drivers. Our method uses a code clone detection technique to extract similarity among device drivers of the same domain. To examine the applicability of our method, we investigated whole device drivers of a Linux source. Results showed that many reusable similar codes can be discerned by the code clone detection method. We also investigated if our method is applicable to other kernel sources. However, the results show that the code clone detection method is not useful for the domain analysis of all kernel sources. That is, the applicability of the code clone detection method to domain analysis is a peculiar feature of device drivers.

• The Journal of Korean Society for School & Community Health Education
• /
• v.2 no.2
• /
• pp.39-55
• /
• 2001

The purpose of this study was to provide the necessary fundamental data in composing a systematic text content of the public health by analyzing each text, domain, and teaching contents suggested in the textbooks as well as teachers' guides of the 7th elementary school education curriculum, while the study subjects were as follows. 1. The health education content suggested in the 7th physical education curriculum were analyzed and examined. 2. The health teaching content of each textbook in the 7th elementary school curriculum was to be analyzed and examined. In order to resolve the above research issues, the physical, spiritual, and social domain along with the (1) Proper living habit, (2) Health and nutrition, (3) Sex education, (4) Prevention of the sense-organic diseases, (5) Cleanliness of food, (6) Oral hygiene, (7) Individual health and public health, (8) Safety in living, (9) Abuse and usage of medication, educational content suggested in the 7tand (10) Environment pollution focused around the health of the elementary school education curriculum was analyzed and its outcome was as below First, compared with the 6th elementary school education curriculum, the health content suggested in the 7th elementary school education curriculum was decreased. Second, although each grade's teaching content of the health domain in the physical education was considered in its structure following after the according systems, they were preponderant in partial subjects such as the safety in living, nutrition, proper living habit, sport, and health in sport. oo. Third, the health education content was organized in 4 units such as the physical growth and development, prevention of diseases, safe living, and leisure living(leisure, spiritual health, and etc.) for the 3rd and 4th grade. Then, as for 5th and 6th grade, it was organized in 3 units such as the understanding the human body, prevention of disease, and leisure and safe living. Fourth, in the physical educational health domain, a strong point was constructed within the physical, spiritual, and social areas of the elementary school physical education. Fifth, the number of the public health education contents directly related with the health education was 43 as with 25 indirect contents. Sixth, each grade's domain unit structure of the public health content was heavy upon the physical and social area throughout every grade while in opposite, the spiritual domain' s unit structure was weak. In according to each grade, the physical domain was stressed in 4, 5, and 6 grades while the social domain was stressed in 1, 5, and 6 grades.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.20 no.1
• /
• pp.129-134
• /
• 2010

In this paper, we propose a redistribution mechanism of the content that is adapted to devices, which may have different display size and computing capabilities, in home network. The proposed system introduces a mechanism that the encrypted content compressed by H.264/SVC(Scalable Video Coding) scheme which has been standardized recently is provided to the device into a level of content suitable to each device capability. To efficiently superdistribute SVC content, this paper defines three requirements and proposes redistribution mechanism which satisfies these requirements using another licence that it is called 'Ticket'. Our system allows devices to redistribute the content freely in the domain using domain key.

• Journal of the Korea Society of Computer and Information
• /
• v.27 no.7
• /
• pp.27-33
• /
• 2022

In this paper, we propose a novel generator architecture for multi-domain style transfer method not an image to image translation, as a method of generating a styled image by transfering a style to the content image. A latent vector and Gaussian noises are added to the generator of GAN so that a high quality image is generated while considering the characteristics of various data distributions for each domain and preserving the features of the content data. With the generator architecture of the proposed GAN, networks are configured and presented so that the content image can learn the styles for each domain well, and it is applied to the domain composed of images of the four seasons to show the high resolution style transfer results.

• Proceedings of the Korean Society of Agricultural Engineers Conference
• /
• 1998.10a
• /
• pp.407-413
• /
• 1998

Considerable progress has been made in the application of time-domain reflectrometry(TDR) to measurement of soil water content. The TDR provides a means of monitoring the water content of soil over a wide range of values, in the field and in the laboratory The TDR measures the volumetric moisture content of the soil via a bureid sensor(probe). Probes can be buried and monitored remotely and an immediate result can be obtained. In addition to, the results are very reliable.

• Proceedings of the Korean Institute of Building Construction Conference
• /
• 2023.11a
• /
• pp.173-174
• /
• 2023

In this study, unit-water content was measured using a frequency domain reflecometry(FDR) sensor that complements the problems of the existing unit-water content measurement method to evaluate the unit-water content affecting the workability, durability, and quality of high strength concrete. The experiment used the unit-water content of high strength concrete as a variable, and the accuracy and probability distribution of the unit-water content measured through deep learning were analyzed for the output value output through the FDR sensor. In the case of the unit-water content predicted by deep learning analysis, a high accuracy and high distribution of more than 93% were found within the error range of ± 10 kg/m³. In the future, research is needed to secure high reliability by utilizing data obtained through experiments with various variables.

• Journal of Korea Multimedia Society
• /
• v.17 no.4
• /
• pp.433-440
• /
• 2014

Wireless communication offers the flexibility to the node movement at the spatial dimension more than the wire communication not only in IP architecture but also in Content-Centric Networking. Although it gives such advantage, the intra-domain movement of a node especially the data source node affects the communication to the access point node which in the end affects the acceptance ratio of the client node that requests the data packets from the data source node. In this paper, we use time-division access method to maintain the acceptance ratio of the client node as the effect of the intra-domain data source node movement in Content-Centric Networking. The simulation result shows that the acceptance ratio of the client node can be maintained using the time-division access method as long as the interval access time is less than the coherence time.