• Smart Media Journal
• /
• v.8 no.2
• /
• pp.58-65
• /
• 2019

Automatic text summarization is a process of shortening a text document by either extraction or abstraction. The abstraction approach inspired by deep learning methods scaling to a large amount of document is applied in recent work. Abstractive text summarization involves utilizing pre-generated word embedding information. Low-frequent but salient words such as terminologies are seldom included to dictionaries, that are so called, out-of-vocabulary(OOV) problems. OOV deteriorates the performance of Encoder-Decoder model in neural network. In order to address OOV words in abstractive text summarization, we propose a copy mechanism to facilitate copying new words in the target document and generating summary sentences. Different from the previous studies, the proposed approach combines accurate pointing information and selective copy mechanism based on bidirectional RNN and bidirectional LSTM. In addition, neural network gate model to estimate the generation probability and the loss function to optimize the entire abstraction model has been applied. The dataset has been constructed from the collection of abstractions and titles of journal articles. Experimental results demonstrate that both ROUGE-1 (based on word recall) and ROUGE-L (employed longest common subsequence) of the proposed Encoding-Decoding model have been improved to 47.01 and 29.55, respectively.

• Journal of Christian Education in Korea
• /
• v.66
• /
• pp.75-112
• /
• 2021

This study is purposed to prepare students in theological universities so that they contribute to their faith community as well as the broader social community in future through developing a practical course model for education of character formation. Facing the challenge of a high-tech revolution being progressed globally in this period, theological seminaries as the Christian-affiliated institute of higher education see the necessity to nurture their learners' competency of holistic personality comprehending cognitive, affective, behavioral, and more spiritual maturity as well as specialty in each subject matter. The holistic character capability of seminarians is closely related not only to their own self-realization, but also social community's well-being. This research had the following content and results. Chapter 1 brought up a subject to educational needs of the holistic character formation in the theological seminaries to meet the rapidly changing times. Chapter 2 considered the importance of character formation in the curriculum of the universities and theological seminaries in the times of high-tech revolution through the literature reviews related to. Chapter 3 clarified the elements of character education to be placed emphasis on theological seminaries and established the way of course planning. Chapter 4, using a course of Christian social education for illustration, described in order the whole process of course planning comprehending the need of planning, description of course concept, establishment and definition of learning outcomes, organization of learning experience(course content), teaching-learning activities, evaluation of learning outcome, and course improvement, which was visualized with the chart of the process of course design. Also, it provided a practical illustration of the conceptual map, an outline of content and an organization of units, and scope-sequence chart. Finally, chapter 5 gave an syllabus example in an integrative and digestive form as the course plan to be distributed to the students, which is the final result produced through the whole process of course planning. This research is intended not to present a overall curriculum planning of the university-level in the theological seminaries, but to develop a course-level curriculum focused on the formation of character. The author hope that this study would be used as a guideline in developing a course model to nurture the capability of the learners' holistic character.

• KIISE Transactions on Computing Practices
• /
• v.23 no.3
• /
• pp.183-187
• /
• 2017

Smart phone users prefer fast reading and texting. Hence, users frequently use abbreviated sequences of words and phrases. Nowadays, abbreviations are widely used from chat terms to technical terms. Therefore, gathering abbreviations would be helpful to many services, including information retrieval, recommendation system, and so on. However, manually gathering abbreviations needs to much effort and cost. This is because new abbreviations are continuously generated whenever a new material such as a TV program or a phenomenon is made. Thus it is required to generate of abbreviations automatically. To generate Korean abbreviations, the existing methods use the rule-based approach. The rule-based approach has limitations, in that it is unable to generate irregular abbreviations. Another problem is to decide the correct abbreviation among candidate abbreviations generated rules. To address the limitations, we propose a method of generating Korean abbreviations automatically using sequence-to-sequence learning in this paper. The sequence-to-sequence learning can generate irregular abbreviation and does not lead to the problem of deciding correct abbreviation among candidate abbreviations. Accordingly, it is suitable for generating Korean abbreviations. To evaluate the proposed method, we use dataset of two type. As experimental results, we prove that our method is effective for irregular abbreviations.

• The Journal of the Korea Contents Association
• /
• v.11 no.1
• /
• pp.357-370
• /
• 2011

Using strategy group theory, this thesis reviewed the status of program providers analysis and the performances it has made so far, and sought measures to improve its limitations. The constraint of program providers analysis based on existing concept of strategy group is that the strategy group was derived from the statistics, and therefore only applied the characteristics of program provider's channels to the analysis, on account of which a systematic and sophisticated classification as well as generalization of strategy or strategy group were hard to obtain. Moreover, the PP strategy variables used to be selected at the firm level and business level, and in relation with resource and competition scope. In future, more appropriate procedure should be followed to obtain objectivity in selecting variables to avoid controversy over intentionality. The measures in this thesis to improve the study of PP strategy group can be summarized as follows: firstly analysis of variables for strategy group classification should be made to single out key variables which are to be classification criteria. Secondly, variables are to be cross-checked by industry experts to increase generalizability. Thirdly, proxy variables should be sublated, and strategy group model which enables the reflection of subsistent properties of PP industry, and the cognitive perception of the executives(CEO) needs to be established. Fourthly, the concepts of mobility barrier and isolating mechanism should be applied to the classification criteria of strategy group to reveal the gap of performance between different strategy groups. Lastly, chronicle study on PP strategy group should be done to perceive the dynamic changes of PP strategy group.

• Journal of Intelligence and Information Systems
• /
• v.29 no.4
• /
• pp.69-90
• /
• 2023

The purpose of this study is to evaluate the service attributes and consumer reactions of telemedicine apps in South Korea and visualize their differentiation by constructing positioning maps. We crawled 23,219 user reviews of 6 major telemedicine apps in Korea from the Google Play store. Topics were derived by BERTopic modeling, and sentiment scores for each topic were calculated through KoBERT sentiment analysis. As a result, five service characteristics in the application attribute category and three in the medical service category were derived. Based on this, a two-dimensional positioning map was constructed through principal component analysis. This study proposes an objective service evaluation method based on text mining, which has implications. In sum, this study combines empirical statistical methods and text mining techniques based on user review texts of telemedicine apps. It presents a system of service attribute elicitation, sentiment analysis, and product positioning. This can serve as an effective way to objectively diagnose the service quality and consumer responses of telemedicine applications.

• KOREAN JOURNAL OF CROP SCIENCE
• /
• v.46 no.6
• /
• pp.453-458
• /
• 2001

The Rural Development Administration (RDA) of Korea now operates a system called Rice Variety Selection Tests (RVST), which are now being implemented in eight Agricultural Research and Extension Services located in eight province RVST's objective is to provide accurate yield estimates and to select well-adapted varieties to each province. Systematic evaluation of entries included in RVST is a highly important task to select the best-adapted varieties to specific location and to observe the performance of entries across a wide range of test sites within a region. The rice yield data in RVST for ordinary transplanting in Kangwon province during 1997-2000 were analyzed. The experiments were carried out in three replications of a random complete block design with eleven entries across five locations. Additive Main effects and Multiplicative Interaction (AMMI) model was employed to examine the interaction between genotype and environment (G$\times$E) in the biplot form. It was found that genotype variability was as high as 66％, followed by G$\times$E interaction variability, 21％, and variability by environment, 13％. G$\times$E interaction was partitioned into two significant (P＜0.05) principal components. Pattern analysis was used for interpretation on G$\times$E interaction and adaptibility. Major determinants among the meteorological factors on G$\times$E matrix were canopy minimum temperature, minimum relative humidity, sunshine hours, precipitation and mean cloud amount. Odaebyeo, Obongbyeo and Jinbubyeo were relatively stable varieties in all the regions. Furthermore, the most adapted varieties in each region, in terms of productivity, were evaluated.