• Journal of Intelligence and Information Systems
• /
• v.27 no.3
• /
• pp.175-197
• /
• 2021

Recently, with the development of deep learning technology, research on unstructured data analysis is being actively conducted, and it is showing remarkable results in various fields such as classification, summary, and generation. Among various text analysis fields, text classification is the most widely used technology in academia and industry. Text classification includes binary class classification with one label among two classes, multi-class classification with one label among several classes, and multi-label classification with multiple labels among several classes. In particular, multi-label classification requires a different training method from binary class classification and multi-class classification because of the characteristic of having multiple labels. In addition, since the number of labels to be predicted increases as the number of labels and classes increases, there is a limitation in that performance improvement is difficult due to an increase in prediction difficulty. To overcome these limitations, (i) compressing the initially given high-dimensional label space into a low-dimensional latent label space, (ii) after performing training to predict the compressed label, (iii) restoring the predicted label to the high-dimensional original label space, research on label embedding is being actively conducted. Typical label embedding techniques include Principal Label Space Transformation (PLST), Multi-Label Classification via Boolean Matrix Decomposition (MLC-BMaD), and Bayesian Multi-Label Compressed Sensing (BML-CS). However, since these techniques consider only the linear relationship between labels or compress the labels by random transformation, it is difficult to understand the non-linear relationship between labels, so there is a limitation in that it is not possible to create a latent label space sufficiently containing the information of the original label. Recently, there have been increasing attempts to improve performance by applying deep learning technology to label embedding. Label embedding using an autoencoder, a deep learning model that is effective for data compression and restoration, is representative. However, the traditional autoencoder-based label embedding has a limitation in that a large amount of information loss occurs when compressing a high-dimensional label space having a myriad of classes into a low-dimensional latent label space. This can be found in the gradient loss problem that occurs in the backpropagation process of learning. To solve this problem, skip connection was devised, and by adding the input of the layer to the output to prevent gradient loss during backpropagation, efficient learning is possible even when the layer is deep. Skip connection is mainly used for image feature extraction in convolutional neural networks, but studies using skip connection in autoencoder or label embedding process are still lacking. Therefore, in this study, we propose an autoencoder-based label embedding methodology in which skip connections are added to each of the encoder and decoder to form a low-dimensional latent label space that reflects the information of the high-dimensional label space well. In addition, the proposed methodology was applied to actual paper keywords to derive the high-dimensional keyword label space and the low-dimensional latent label space. Using this, we conducted an experiment to predict the compressed keyword vector existing in the latent label space from the paper abstract and to evaluate the multi-label classification by restoring the predicted keyword vector back to the original label space. As a result, the accuracy, precision, recall, and F1 score used as performance indicators showed far superior performance in multi-label classification based on the proposed methodology compared to traditional multi-label classification methods. This can be seen that the low-dimensional latent label space derived through the proposed methodology well reflected the information of the high-dimensional label space, which ultimately led to the improvement of the performance of the multi-label classification itself. In addition, the utility of the proposed methodology was identified by comparing the performance of the proposed methodology according to the domain characteristics and the number of dimensions of the latent label space.

• Korean Journal of Heritage: History & Science
• /
• v.53 no.4
• /
• pp.198-215
• /
• 2020

Official development assistance refers to assistance provided by governments and other public institutions in donor countries, aimed at promoting economic development and social welfare in developing countries. The purpose of this research is to examine the construction process of the "Myanmar Cultural Heritage Management System" that is underway as part of the ODA project to strengthen cultural and artistic capabilities and analyze the achievements and challenges of the Digital Cultural Heritage ODA. The digital cultural heritage management system is intended to achieve the permanent preservation and sustainable utilization of tangible and intangible cultural heritage materials. Cultural heritage can be stored in digital archives, newly approached using computer analysis technology, and information can be used in multiple dimensions. First, the Digital Cultural Heritage ODA was able to permanently preserve cultural heritage content that urgently needed digitalization by overcoming and documenting the "risk" associated with cultural heritage under threat of being extinguished, damaged, degraded, or distorted in Myanmar. Second, information on Myanmar's cultural heritage can be systematically managed and used in many ways through linkages between materials. Third, cultural maps can be implemented that are based on accurate geographical location information as to where cultural heritage is located or inherited. Various items of cultural heritage were collectively and intensively visualized to maximize utility and convenience for academic, policy, and practical purposes. Fourth, we were able to overcome the one-sided limitations of cultural ODA in relations between donor and recipient countries. Fifth, the capacity building program run by officials in charge of the beneficiary country, which could be the most important form of sustainable development in the cultural ODA, was operated together. Sixth, there is an implication that it is an ODA that can be relatively smooth and non-face-to-face in nature, without requiring the movement of manpower between countries during the current global pandemic. However, the following tasks remain to be solved through active discussion and deliberation in the future. First, the content of the data uploaded to the system should be verified. Second, to preserve digital cultural heritage, it must be protected from various threats. For example, it is necessary to train local experts to prepare for errors caused by computer viruses, stored data, or operating systems. Third, due to the nature of the rapidly changing environment of computer technology, measures should also be discussed to address the problems that tend to follow when new versions and programs are developed after the end of the ODA project, or when developers have not continued to manage their programs. Fourth, since the classification system criteria and decisions regarding whether the data will be disclosed or not are set according to Myanmar's political judgment, it is necessary to let the beneficiary country understand the ultimate purpose of the cultural ODA project.

• Economic and Environmental Geology
• /
• v.54 no.1
• /
• pp.77-89
• /
• 2021

In this study, a method has been proposed to improve the performance of hydraulic property estimation model developed by Jeong et al. (2020). In their study, low-dimensional features of the annual groundwater level (GWL) fluctuation patterns extracted based on a Denoising autoencoder (DAE) was used to develop a regression model for predicting hydraulic properties of an aquifer. However, low-dimensional features of the DAE are highly dependent on the precipitation pattern even if the GWL is monitored at the same location, causing uncertainty in hydraulic property estimation of the regression model. To solve the above problem, a process for generating the GWL fluctuation pattern for conditioning the precipitation is proposed based on a conditional variational autoencoder (CVAE). The CVAE trains a statistical relationship between GWL fluctuation and precipitation pattern. The actual GWL and precipitation data monitored on a total of 71 monitoring stations over 10 years in South Korea was applied to validate the effect of using CVAE. As a result, the trained CVAE model reasonably generated GWL fluctuation pattern with the conditioning of various precipitation patterns for all the monitoring locations. Based on the trained CVAE model, the low-dimensional features of the GWL fluctuation pattern without interference of different precipitation patterns were extracted for all monitoring stations, and they were compared to the features extracted based on the DAE. Consequently, it can be confirmed that the statistical consistency of the features extracted using CVAE is improved compared to DAE. Thus, we conclude that the proposed method may be useful in extracting a more accurate feature of GWL fluctuation pattern affected solely by hydraulic characteristics of the aquifer, which would be followed by the improved performance of the previously developed regression model.

• Journal of Korean Historical Folklife
• /
• no.27
• /
• pp.267-361
• /
• 2008

This article traces how the modern Chinese "nation" was constructed as an "imagined community" around Huang-ti (the Yellow Emperor) in late Qing. Huang-ti was a legendary figure in ancient China and the imperial courts monopolized the worship of him. Many late Qing intellectuals appropriated this symbolic figure and, through a set of discursive strategies of "framing, voice and narrative structure," transformed him into a privileged symbol for modern Chinese national identity. What Huang-ti could offer was, however, no more than a "public face" for the imagined new national community, or in other words, a formal structure without substantial contents. No consensus appeared on whom the Chinese nation should include and where the Chinese nation should draw its boundaries. The anti-Manchu revolutionaries emphasized the primordial attachment of blood and considered modern China an exclusive community of Huang-ti's descent. The constitutional reformers sought to stretch the boundaries to include the ethnic groups other than the Han. Some minority intellectuals, particularly the Manchu ones, re-constructed the historic memory of their ethnic origin around Huang-ti. The quarrels among intellectuals of different political persuasion testify how Huang-ti as the most powerful cultural symbol became a site for contests and negotiations in the late Qing process of national construction.

• Asia-Pacific Journal of Business Venturing and Entrepreneurship
• /
• v.15 no.6
• /
• pp.27-42
• /
• 2020

Startup accelerators have emerged as new investment entities that help early startups, which are not easy to survive continuously due to lack of funds, commercialization capabilities, and experiences. As their positive performance on early startups and the ecosystem has been proven, the number of early startups which want to receive their investment is also increasing. However, they are vaguely preparing to attract accelerators' investment because they do not have any information on what factors the accelerators consider important. In addition, researches on startup accelerators are also at an early level, so there are no remarkable prior studies on factors that decide on investment. Therefore, this study aims to help startups prepare for investment attraction by looking at what factors are important for accelerators to invest, and to provide meaningful implications to academia. In the preceding study, we derived five upper level categories, 26 lower level accelerators' investment determinants through the qualitative meta-synthesis method, secondary data analysis, observation on US accelerators and in-depth interviews. In this study, we want to derive important implications by deriving priorities of the accelerators' investment determinants. Therefore, we used AHP that are evaluated as the suitable methodology for deriving importance and priority. The analysis results show that accelerators value market-related factors most. This means that startups that are subject to investment by accelerators are early-stage startups, and many companies have not fully developed their products or services. Therefore, market-related factors that can be evaluated objectively seem to be more important than products (or services) that are still ambiguous. Next, it was found that the factors related to the internal workforce of startups are more important. Since accelerators want to develop their businesses together with start-ups and team members through mentoring, ease of collaboration with them is very important, which seems to be important. The overall priority analysis results of the 26 investment determinants show that 'customer needs' and 'founders and team members' understanding of customers and markets' (0.62) are important and high priority factors. The results also show that startup accelerators consider the customer-centered perspective very important. And among the factors related to startups, the most prominent factor was the founder's openness and execution ability. Therefore, it can be confirmed that accelerators consider the ease of collaboration with these startups very important.

• Journal of Intelligence and Information Systems
• /
• v.27 no.4
• /
• pp.49-71
• /
• 2021

Users who intend to utilize knowledge to actively solve given problems proceed their jobs with cross- and sequential exploration of associated knowledge related each other in terms of certain criteria, such as content relevance. A knowledge map is the diagram or taxonomy overviewing status of currently managed knowledge in a knowledge-base, and supports users' knowledge exploration based on certain relationships between knowledge. A knowledge map, therefore, must be expressed in a networked form by linking related knowledge based on certain types of relationships, and should be implemented by deploying proper technologies or tools specialized in defining and inferring them. To meet this end, this study suggests a methodology for developing the knowledge graph-based knowledge map using the Graph DB known to exhibit proper functionality in expressing and inferring relationships between entities and their relationships stored in a knowledge-base. Procedures of the proposed methodology are modeling graph data, creating nodes, properties, relationships, and composing knowledge networks by combining identified links between knowledge. Among various Graph DBs, the Neo4j is used in this study for its high credibility and applicability through wide and various application cases. To examine the validity of the proposed methodology, a knowledge graph-based knowledge map is implemented deploying the Graph DB, and a performance comparison test is performed, by applying previous research's data to check whether this study's knowledge map can yield the same level of performance as the previous one did. Previous research's case is concerned with building a process-based knowledge map using the ontology technology, which identifies links between related knowledge based on the sequences of tasks producing or being activated by knowledge. In other words, since a task not only is activated by knowledge as an input but also produces knowledge as an output, input and output knowledge are linked as a flow by the task. Also since a business process is composed of affiliated tasks to fulfill the purpose of the process, the knowledge networks within a business process can be concluded by the sequences of the tasks composing the process. Therefore, using the Neo4j, considered process, task, and knowledge as well as the relationships among them are defined as nodes and relationships so that knowledge links can be identified based on the sequences of tasks. The resultant knowledge network by aggregating identified knowledge links is the knowledge map equipping functionality as a knowledge graph, and therefore its performance needs to be tested whether it meets the level of previous research's validation results. The performance test examines two aspects, the correctness of knowledge links and the possibility of inferring new types of knowledge: the former is examined using 7 questions, and the latter is checked by extracting two new-typed knowledge. As a result, the knowledge map constructed through the proposed methodology has showed the same level of performance as the previous one, and processed knowledge definition as well as knowledge relationship inference in a more efficient manner. Furthermore, comparing to the previous research's ontology-based approach, this study's Graph DB-based approach has also showed more beneficial functionality in intensively managing only the knowledge of interest, dynamically defining knowledge and relationships by reflecting various meanings from situations to purposes, agilely inferring knowledge and relationships through Cypher-based query, and easily creating a new relationship by aggregating existing ones, etc. This study's artifacts can be applied to implement the user-friendly function of knowledge exploration reflecting user's cognitive process toward associated knowledge, and can further underpin the development of an intelligent knowledge-base expanding autonomously through the discovery of new knowledge and their relationships by inference. This study, moreover than these, has an instant effect on implementing the networked knowledge map essential to satisfying contemporary users eagerly excavating the way to find proper knowledge to use.

• Journal of Intelligence and Information Systems
• /
• v.28 no.2
• /
• pp.191-206
• /
• 2022

Recently, it is a de-facto approach to utilize a pre-trained language model(PLM) to achieve the state-of-the-art performance for various natural language tasks(called downstream tasks) such as sentiment analysis and question answering. However, similar to any other machine learning method, PLM tends to depend on the data distribution seen during the training phase and shows worse performance on the unseen (Out-of-Distribution) domain. Due to the aforementioned reason, there have been many efforts to develop domain-specified PLM for various fields such as medical and legal industries. In this paper, we discuss the training of a finance domain-specified PLM for the Korean language and its applications. Our finance domain-specified PLM, KB-BERT, is trained on a carefully curated financial corpus that includes domain-specific documents such as financial reports. We provide extensive performance evaluation results on three natural language tasks, topic classification, sentiment analysis, and question answering. Compared to the state-of-the-art Korean PLM models such as KoELECTRA and KLUE-RoBERTa, KB-BERT shows comparable performance on general datasets based on common corpora like Wikipedia and news articles. Moreover, KB-BERT outperforms compared models on finance domain datasets that require finance-specific knowledge to solve given problems.

• Conservation Science in Museum
• /
• v.27
• /
• pp.147-164
• /
• 2022

Over 200,000 Korean cultural heritage items are currently located abroad. They have made their way to 22 countries under different circumstances and with unique backgrounds. While some of them continue to contribute to promoting Korean culture around the world, others cannot be exhibited due to damage or poor condition. In view of these circumstances, the Overseas Korean Cultural Heritage Foundation (OKCHF) has since 2013 provided museums and art galleries abroad with support for the conservation, restoration, and utilization of the Korean cultural heritage items that they house. As a part of these efforts and on the occasion of the 120th anniversary of the diplomatic relationship between the Republic of Korea and the Kingdom of Belgium in 2021, a gilt-bronze case for acupuncture needles dating to the Goryeo period (918-1392) from the collection of the Royal Museums of Art and History (RMAH), Belgium was brought to Korea for conservation treatment. The primary purpose of this conservation treatment was to restore the original form of the relic and slow to the degree possible the progress of corrosion. The conservation treatment of the gilt-bronze case followed the fundamental order of conservation treatment: removal of corrosive substances, stabilization, and reinforcement. Since this was the first case of restoring metallic cultural properties under the abovementioned support program by the OKCHF, special methodologies distinct from those available in overseas institutions were required. Diverse scientific methods (e.g., X-ray inspection, CT scanning, 3D microscopy) were applied to identify the metalcraft techniques used in the Goryeo period. The analysis found that several designs, including lotus and scrollwork, were exquisitely engraved on the surface of the case by making dots using a round-edged chisel. A bronze plate engraved with designs was rolled into a cylindrical form. The ends were overlapped by 2 to 3 centimeters and then attached to each other by silver soldering. The overlapping ends were welded flat with nearly no gaps. As the final process in the production, the case was lavishly gilt with gold powder using amalgam gilding. The conservation treatment of the gilt-bronze case for acupunctural needles in the RMAH collection restored the original form of the relic and arrested further corrosion. Above all, it revived the historic and academic value of the overseas Korean cultural heritage through scientific analysis.

• Asia-Pacific Journal of Business Venturing and Entrepreneurship
• /
• v.16 no.6
• /
• pp.131-144
• /
• 2021

The organizational ambidexterity is an organizational strategy designed to pursue exploration activities to seize new opportunities and exploitation activities to efficiently use resources. Most of these ambidextrous structures have been studied for large corporations with slack resources, and there are still not many studies on the necessity of an ambidextrous structure for startups with relatively low-level resources. However, recently, the startup ecosystem is being advanced globally, and the amount of VC investment is rapidly increasing. This is a time when a lot of venture fund is invested in startups and a startup-friendly environment for rapid growth is created. This is the time to discuss the necessity and applicability of an ambidextrous organizational structure for startups. Therefore, this study conducted a hypothesis test whether the importance and necessity of balance that startups solving market problems with new ideas and utilizing accumulated resources have. To conduct this study, we analyzed 140 startups data gathered from the survey and the moderation effect was also analyzed. As a result of the study, it was verified that the balance of startup exploration and exploitation had a significant effect on startup performance, and the moderating effect of environmental dynamics was found to have a significant effect on the relationship with non-financial performance. Therefore, for startups with insufficient resources, it was concluded that the surplus resources generated in the process of a firm's growth should be effectively utilized and the balance between exploration and exploitation should be balanced from the initial stage of searching for a new business. In other words, it was confirmed that it is important for continuous growth and survival to seek the structure of an ambidextrous organization in order to internalize a mechanism that enables startups to pursue both effectiveness and efficiency in the long term. This study suggests a strategic direction for the growth of startups from the perspective of organizational structure. We expect that this meaningful results on the relationship between the ambidextrous capabilities of startups and performance contribute to the growth of startups in the rapidly growing startup venture environment.

• Journal of Appropriate Technology
• /
• v.7 no.2
• /
• pp.136-143
• /
• 2021

Pacific island countries, including Kiribati, are suffering from a shortage of essential resources as well as a reduction in their living space due to sea level rise and coastal erosion from climate change, groundwater pollution and vegetation changes. Global activities to solve these problems are being progressed by the UN's efforts to implement SDGs. Pacific island countries can adapt to climate change by using abundant marine resources. In other words, seawater plants can assist in achieving SDGs #2, #6 and #7 based on SDGs #14 in these Pacific island countries. Under the auspice of Korea International Cooperation Agency (KOICA), Korea Research Institute of Ships and Ocean Engineering (KRISO) established the Sustainable Seawater Utilization Academy (SSUA) in 2016, and its 30 graduates formed the SSUA Kiribati Association in 2017. The Ministry of Oceans and Fisheries (MOF) of the Republic of Korea awarded ODA fund to the Association. By taking advantage of seawater resource and related plants, it was able to provide drinking water and vegetables to the local community from 2018 to 2020. Among the various fields of education and practice provided by SSUA, the Association hope to realize hydroponic cultivation and seawater desalination as a self-support project through a pilot project. To this end, more than 140 households are benefiting from 3-stage hydroponics, and a seawater desalination system in connection with solar power generation was installed for operation. The Association grows and supplies vegetable seedlings from the provided seedling cultivation equipment, and is preparing to convert to self-support business from next year. The satisfaction survey shows that Tarawa residents have a high degree of satisfaction with the technical support and its benefits. In the future, it is hoped that SSUA and regional associations will be distributed to neighboring island countries to support their SDGs implementations.