• Smart Media Journal
• /
• v.11 no.4
• /
• pp.30-37
• /
• 2022

Disease threatens plant growth and recognizing the type of disease is essential to making a remedy. In recent years, deep learning has witnessed a significant improvement for this task, however, a large volume of labeled images is one of the requirements to get decent performance. But annotated images are difficult and expensive to obtain in the agricultural field. Therefore, designing an efficient and effective strategy is one of the challenges in this area with few labeled data. Transfer learning, assuming taking knowledge from a source domain to a target domain, is borrowed to address this issue and observed comparable results. However, current transfer learning strategies can be regarded as a supervised method as it hypothesizes that there are many labeled images in a source domain. In contrast, unsupervised transfer learning, using only images in a source domain, gives more convenience as collecting images is much easier than annotating. In this paper, we leverage unsupervised transfer learning to perform plant disease recognition, by which we achieve a better performance than supervised transfer learning in many cases. Besides, a vision transformer with a bigger model capacity than convolution is utilized to have a better-pretrained feature space. With the vision transformer-based unsupervised transfer learning, we achieve better results than current works in two datasets. Especially, we obtain 97.3% accuracy with only 30 training images for each class in the Plant Village dataset. We hope that our work can encourage the community to pay attention to vision transformer-based unsupervised transfer learning in the agricultural field when with few labeled images.

• Journal of Korea Society of Digital Industry and Information Management
• /
• v.19 no.2
• /
• pp.27-37
• /
• 2023

The purpose of this study was to analyze the effect of the online extracurricular program operated by the university. The program contents applied in the study included learning strategies such as time management, goal setting, note taking, and memorization methods. The program used in the study was operated in an online environment, and the content was developed between 27 and 29 minutes. The developed contents can be taken using the learning management system. The variables selected to analyze the effects of this program were learning strategies and learning flow, and satisfaction was also included to examine the responses of program participants. The results of the study are as follows. First, learning strategies and learning flow showed statistically significant differences. This result is because the content was composed of meaningful sub-topics by selecting the elements necessary for learning activities. Second, as a result of program satisfaction analysis, it was confirmed that the average for all questions was high. Among them, the average of the item that the theme and contents of the program were useful was the highest. Third, open responses were analyzed by classifying them into cognitive and affective domains. In the cognitive domain, meanings such as knowledge, understanding, and application were presented as keywords, and in the affective domain, a number of keywords for motivation and attitude change were presented. This study is significant in that it provided practical programs necessary for university freshmen and analyzed their effects.

• Convergence Security Journal
• /
• v.14 no.7
• /
• pp.75-84
• /
• 2014

The advancement in ubiquitous healthcare specifically in preventive healthcare can lead to longer life expectancy especially for the elderly patients. To aid in preventing premature loss of lives as well as lengthening life span, this research aims to implement the use of mobile and wireless sensor technology to improve the quality of life and lengthen life expectancy. The threats to privacy and security have received increasing attention as ubiquitous healthcare applications over the Internet become more prevalent, mobile and universal. Therefore, we propose Context-aware Service of U-Healthcare Application based Knowledge using Ontology in secure health information exchange. This research also applies ontology in secure information exchange to support knowledge base, context modeling, and context reasoning by applying the general application areas for ontologies to the domain of context in ubiquitous computing environments. This paper also demonstrates how knowledge base, context technologies, and mobile web services can help enhance the quality of services in preventive ubiquitous healthcare to elderly patients.

• Journal of Korean Academy of Fundamentals of Nursing
• /
• v.14 no.2
• /
• pp.150-156
• /
• 2007

Purpose: This study was conducted to identify the nature the research trends of articles published in the Journal of Korean Academy of Fundamentals of Nursing (JKAFN). Method: Comparison analysis between articles in JKAFN (Volume 13, Number 1-3) and articles in the Journal of Korean Academy of Nursing (Volume 36, Number 1-3, 5-7) based on knowledge development classification was used to identify the nature of research trends in JKAFN. Results: Based on comparison with Journal of Korean Academy of Nursing, research trends in JKAFN were dominant; in personal knowledge in the pattern of knowing in nursing, in the desiderative focus in cognitive needs for nursing epistemology, in advancing of various aspects of nursing in the focus of knowledge, and practice domain in the domains for nursing. Conclusion: The major focuses on research trends in JKAFN were nursing practice, actions of nurses in practice and practice guideline applicable in nursing education.

• Hwankyungkyoyuk
• /
• v.12 no.1
• /
• pp.40-63
• /
• 1999

As the environmental problems are recognized as daily problems in our lives, not as issues of those who are engaged in specific professional fields, the interest in environmental education is increasing gradually. The environmental education is the one that studies the environment and its problems and seeks the solutions for them. This paper deals with how the Korean subject will include environmental education. The environmental problems are already dealt with in the Korean textbooks according to the 6th curriculum for the Korean subject. A noticeable fact is that those textbooks connect the skills for language skills with environmental education. That is, the textbooks try to improve 4 language skills (speaking, listening, reading, and writing) with the Korean data related to environment, which can be the most practical means. This tendency will be also reflected in the 7th curriculum for the Korean subject, and the means will be taken by which environmental education will be able to be implemented more effectively through a variety of learning activities. In case of speaking and listening, learning activities such as speaking of, listening to, or discussing the contents concerning environmental problems can be recommended. In case of reading and literature, learning activities such as reading articles or works concerning environmental problems. Through these learning activities the Korean education will be able to achieve the goal in the fields of knowledge, information, and autonomy or attitudes which are the goals of environmental education. If the contents of the Korean curriculum are described in detail, it can be known that the Korean subject have someting to do with knowledge, skills, and recognition more deeply. In the methods In obtain information and knowledge, it will be desirable to recognize knowledge and information indirectly through various reading data rather than to recognize knowledge and information directly. Or it will be desirable to increase the sensitivity about environmental problems through literary works. For this environmental education in the above, we need to utilize discussion or presentation-oriented leaching and learning in the Korean education. Also we need to approach environmental problems by using various teaching media. We need to emphasize the education in the affective domain, especially through expression of emotions. guidance of reactions, internalization, personification, and so on.

• Journal of the Ergonomics Society of Korea
• /
• v.25 no.3
• /
• pp.49-60
• /
• 2006

Technological encroachment provides human operators with flood of information that must be analyzed to understand the environment and make judgments that lead to strategic actions. Further, the environment is not static and therefore uncertain, changing its aspect dynamically. Complexity accompanied with its dynamics imposes substantial difficulty to human operators' task. Criticality of having situational understanding becomes more important than ever. Situationalunderstanding requires the human operators possessing tacit knowledge in order for them to make the sense out of the situation while interacting with information from many heterogeneous sources, the notion of sensemaking. Sensemaking refers to the process of developing mental framework to assemble pieces of information representing different aspects of the environment that can be used to develop one's own actionable knowledge to implement their judgments in the uncertain environment. Therefore, judgment process and performance is a key component of sensemaking process. Among many judgment and decision making models, the lens model with its extension can be utilized to partially describe the judgmental aspect of sensemaking. One of the lens model parameters, unmodeled knowledge, can be a corresponding quantitative measure for the tacit knowledge that plays an important role in sensemaking. In this paper, a comprehensive literature for sensemaking is provided to formally define the notion of sensemaking in the military domain. Also, it is proposed that there is a crucial link between the sensemaking and human judgment process and performance from the lens model perspective. Potential implications for experimental framework are also proposed.

• The Journal of Information Systems
• /
• v.16 no.4
• /
• pp.131-155
• /
• 2007

The purpose of this thesis is to study on the philosophical analysis model and its methodological application of information systems research evaluation from critical realist perspective. Fist of all, I examine ontological epistemological methodological assertions of critical realism. Because the philosophy of critical realism is an opportunity for information systems study. I examine Dobson and Mutch's critical realist perspective on actors-structure model. I suggest a critical realist actors-praxis-structure model. This model provides the potential for a new approach to social investigations in its provision of an ontology for the analytical separation of structure and agency. Of most importance might be the incorporation of non-humans into the analysis of social interaction and of technology into the elaboration of structures. I also examine Tsoukas's critical realistic meta-theory of management. I suggest a critical realist IS management model. This model elucidate the nature of management and delineate the scope of applicability of various perspectives on management. The causal powers of management reside in the real domain and, taken together, their logics are contradictory, the effects of their contradictory composition are contingent upon prevailing contingencies. I analyze Carlsson's theory of design knowledge. His framework builds on that the aim of IS design science research is to develop practical knowledge for the design and realization of different classes of IS initiatives, where IS are viewed as socio-technical systems and not just IT artefacts. The framework proposes that the output of IS design science research is practical IS design knowledge in the form of field-tested and grounded technological rules. The IS design knowledge is developed through an IS design science research cycle. In conclusion, I think that IS actors-praxis-structure model, meta-theoretical IS management model, and IS design knowledge model according to critical realistic approach are very useful for IS research evaluation. Nevertheless, important problems are left not resolved.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.31 no.8B
• /
• pp.716-729
• /
• 2006

The information systems in the most enterprise environments are distributed locally and are comprised with various heterogeneous data sources, so that it is difficult to obtain necessary and integrated information for supporting user decision. For solving 'this problems efficiently, it provides uniform interface to users and constructed database systems between heterogeneous systems make a consistence each independence and need to provide transparency like one interface. This paper presents XMDR that consists of category, standard ontology, location ontology and knowledge base. Standard ontology solves heterogeneous problem about naming, attributes, relations in data expression. Location ontology is a mediator that connects each legacy systems. Knowledge base defines the relation for sharing glossary. Adaptive retrieve proposes integrated retrieve system through reflecting site weight by location ontology, information sharing of various forms of knowledge base and integration and propose conceptual domain model about how to share unstructured knowledge.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.17 no.4
• /
• pp.275-287
• /
• 2024

Research on the integration of Knowledge Graphs (KGs) and Language Models (LMs) has been consistently explored over the years. However, studies focusing on the automatic generation of text using the structured knowledge from KGs have not been as widely developed. In this study, we propose a methodology for automatically generating specific domain-related research items (Related Work) at a level comparable to existing papers. This methodology involves: 1) selecting optimal prompts, 2) extracting triples through a four-step refinement process, 3) constructing a knowledge graph, and 4) automatically generating related research. The proposed approach utilizes GPT-4, one of the large language models (LLMs), and is desigend to automatically generate related research by applying the four-step refinement process. The model demonstrated performance metrics of 17.3, 14.1, and 4.2 in Triple extraction across #Supp, #Cont, and Fluency, respectively. According to the GPT-4 automatic evaluation criteria, the model's performamce improved from 88.5 points vefore refinement to 96.5 points agter refinement out of 100, indicating a significant capability to automatically generate related research at a level similar to that of existing papers.

• Journal of Intelligence and Information Systems
• /
• v.28 no.4
• /
• pp.251-273
• /
• 2022

Recently, research on applying text analysis to deep learning has steadily continued. In particular, researches have been actively conducted to understand the meaning of words and perform tasks such as summarization and sentiment classification through a pre-trained language model that learns large datasets. However, existing pre-trained language models show limitations in that they do not understand specific domains well. Therefore, in recent years, the flow of research has shifted toward creating a language model specialized for a particular domain. Domain-specific pre-trained language models allow the model to understand the knowledge of a particular domain better and reveal performance improvements on various tasks in the field. However, domain-specific further pre-training is expensive to acquire corpus data of the target domain. Furthermore, many cases have reported that performance improvement after further pre-training is insignificant in some domains. As such, it is difficult to decide to develop a domain-specific pre-trained language model, while it is not clear whether the performance will be improved dramatically. In this paper, we present a way to proactively check the expected performance improvement by further pre-training in a domain before actually performing further pre-training. Specifically, after selecting three domains, we measured the increase in classification accuracy through further pre-training in each domain. We also developed and presented new indicators to estimate the specificity of the domain based on the normalized frequency of the keywords used in each domain. Finally, we conducted classification using a pre-trained language model and a domain-specific pre-trained language model of three domains. As a result, we confirmed that the higher the domain specificity index, the higher the performance improvement through further pre-training.