• Proceedings of the Korea Water Resources Association Conference
• /
• 2022.05a
• /
• pp.329-329
• /
• 2022

Accurate characterization of terrestrial precipitation variation from high spatial resolution satellite sensors is beneficial for urban hydrology and microscale agriculture modeling, as well as natural disasters (e.g., urban flooding) early warning. However, the widely-used top-down approach for precipitation retrieval from microwave satellites is limited in several hydrological and agricultural applications due to their coarse spatial resolution. In this research, we aim to apply a novel bottom-up method, the parameterized SM2RAIN, where precipitation can be estimated from soil moisture signals based on an inversion of water balance model, to generate high spatial resolution terrestrial precipitation estimates at 0.01º grid (roughly 1-km) from the C-band SAR Sentinel-1. This product was then tested against a common reanalysis-based precipitation data and a domestic rain gauge network from the Korean Meteorological Administration (KMA) over central South Korea, since a clear difference between climatic types (coasts and mainlands) and land covers (croplands and mixed forests) was reported in this area. The results showed that seasonal precipitation variability strongly affected the SM2RAIN performances, and the product derived from separated parameters (rainy and non-rainy seasons) outperformed that estimated considering the entire year. In addition, the product retrieved over the mainland mixed forest region showed slightly superior performance compared to that over the coastal cropland region, suggesting that the 6-day time resolution of S1 data is suitable for capturing the stable precipitation pattern in mainland mixed forests rather than the highly variable precipitation pattern in coastal croplands. Future studies suggest comparing this product to the traditional top-down products, as well as evaluating their integration for enhancing high spatial resolution precipitation over entire South Korea.

• The Transactions of the Korea Information Processing Society
• /
• v.3 no.7
• /
• pp.1812-1821
• /
• 1996

In this paper, firstly we describe the structure and the performance of our ICPS(Information Communicaion Processing System) which currently provides information retrieval services, and then make a proposal for the construction of the open-networking information communication infra-structure which enables us to fully pre-pare for the emerging information society. In detail, the structure and the methodology needed for the implementation of the billing function on behalf of all information providers by using the user access network number as a user identification number while guaranteeing the equivalent access to the multiple value-added networks, are suggested. Based on the above ideas, the AICPS(Advanced Information Communication Processing System) has been designed and implemented. Final system performance evaluation with the assumption of a poling system as a system model, shows that our system can handle 10,000 user simultaneously who are using V.34 28.8 kbps modems and the processing capacity is 288,000 packet/sec. This result is so far superior to our target performance established during the desingning procedure. Namely, our system was originally designed to accommodate only 960 users at the same time. By taking advantage of this excessive high performance of our system, many other users can easily access the new services which are accessible only throught the ISDN or the Internet.

• Journal of the Korean Society for information Management
• /
• v.25 no.2
• /
• pp.183-198
• /
• 2008

With the purpose of enriching existing catalogues with FRBR, which is the Functional Requirements for Bibliographic Records, in mind, this paper aims to evaluate the impact of bibliographic ontology on the overall system's performance in the field of literature. In doing this, OCLC's FictionFinder(http://fictionfinder.oclc.org) was selected and qualitatively evaluated. In this study 40 university seniors evaluated the following three aspects using the 'transferring thoughts onto paper method': 1) In which ways is this FRBR-aware bibliographical ontology helpful? 2) Are the things which are initially attempted to be helped being helped? 3) Would users seeking one work in particular also see all other related works? In conclusion, this study revealed that, as Cutter claimed in his $2^{nd}$ rule of the library, collocations give added-value to the users and overall ontology provides better interface and usefulness. It also revealed that a system's evaluation with qualitative methodology helped to build full pictures of the system and to grip the information needs of the users when the system is developed. Qualitative evaluations, therefore, could be used as indicators for the evaluation of any information retrieval systems.

• KIPS Transactions on Software and Data Engineering
• /
• v.2 no.8
• /
• pp.535-542
• /
• 2013

Since case-based reasoning(CBR) has many advantages, it has been used for supporting decision making in various areas including medical checkup, production planning, customer classification, and so on. However, there are several factors to be set by heuristics when designing effective CBR systems. Among these factors, this study addresses the issue of selecting appropriate neighbors in case retrieval step. As the criterion for selecting appropriate neighbors, conventional studies have used the preset number of neighbors to combine(i.e. k of k-nearest neighbor), or the relative portion of the maximum similarity. However, this study proposes to use the absolute similarity threshold varying from 0 to 1, as the criterion for selecting appropriate neighbors to combine. In this case, too small similarity threshold value may make the model rarely produce the solution. To avoid this, we propose to adopt the coverage, which implies the ratio of the cases in which solutions are produced over the total number of the training cases, and to set it as the constraint when optimizing the similarity threshold. To validate the usefulness of the proposed model, we applied it to a real-world target marketing case of an online shopping mall in Korea. As a result, we found that the proposed model might significantly improve the performance of CBR.

• Journal of the Korea Convergence Society
• /
• v.12 no.12
• /
• pp.93-106
• /
• 2021

Blenderbot 2.0 is a dialogue model representing open domain chatbots by reflecting real-time information and remembering user information for a long time through an internet search module and multi-session. Nevertheless, the model still has many improvements. Therefore, this paper analyzes the limitations and errors of BlenderBot 2.0 from three perspectives: model, data, and dialogue. From the data point of view, we point out errors that the guidelines provided to workers during the crowdsourcing process were not clear, and the process of refining hate speech in the collected data and verifying the accuracy of internet-based information was lacking. Finally, from the viewpoint of dialogue, nine types of problems found during conversation and their causes are thoroughly analyzed. Furthermore, practical improvement methods are proposed for each point of view, and we discuss several potential future research directions.

• Journal of the Korean Society for Library and Information Science
• /
• v.46 no.2
• /
• pp.55-82
• /
• 2012

This study seeks to develop a curriculum operating model for Korean Library and Information Science, based on investigations into LIS curricula at home and abroad. Standard courses that can be applied to this model were also proposed. This study comprehensively analyzed the contents of domestic and foreign curricula and surveyed current librarians in all types of library fields. As a result, this study proposed required courses, core courses, and elective courses. Six required LIS courses are: Introduction to Library and Information Science, Information Organization, Information Services, Library and Information Center Management, Information Retrieval, and Field Work. Six core LIS courses are: Classification & Cataloging Practice, Subject Information Resources, Collection Development, Digital Library, Introduction to Bibliography, and Introduction to Archive Management. Twenty selective LIS courses include: the General Library and Information Science area (Cultural History of Information, Information Society and Library, Library and Copyright, Research Methods in Library and Information Science), the Information Organization area (Metadata Fundamentals, KORMARC Practice), the Information Services area (Information Literacy Instruction, Reading Guidance, Information User Study), the Library and Information Center Management area (Library Management, including management for different kinds of libraries, Library Information Cooperator, Library Marketing, Non-book Material and Multimedia Management (Contents Management), the Information Science area (Database Management, including Web DB Management, Indexing and Abstracting, Introduction to Information Science, Understanding Information Science, Automated System of Library, Library Information Network), and the Archival Science area (Preservation Management).

• Journal of the Korean earth science society
• /
• v.32 no.1
• /
• pp.21-32
• /
• 2011

Using six-year (2004-2009) SKYNET measurements, MODIS-derived AOTs were validated at five SKYNET sites (Seoul, Chiba, Etchujima, Fukuejima, and Hedomisaki), in addition to climatological analysis of MODIS-derived optical properties over the East Asian domain ($20-50^{\circ}N$, $90-150^{\circ}E$). In so doing MODIS-SKYNET collocated AOT data were constructed if two measurements are taken within 25 km distance and within 30 minute time difference. From the comparison of two measurements, it is demonstrated that aerosol type insignificantly affects the accuracy of MODIS AOT. It is because the aerosol model combining predefined fine aerosol model and coarse aerosol model is used for the retrieval. However, positive bias between MODIS and SKYNET increases as fraction of the coarse aerosol model increases. In addition, MODIS AOT appears to be overestimated in case of lower aerosol loading while the overestimation tends to decrease with increased aerosol loading. Regression analysis between MODIS AOT and SKYNET AOT for 550 nm band yields 0.86, 0.16, and 0.61 of regression slope, intercept, and coefficient of determination, respectively. Those statistical results may draw a conclusion that MODIS AOTs over East Asia carry a reasonable accuracy compared to ground-based SKYNET measurements.

• Journal of Korea Water Resources Association
• /
• v.39 no.10 s.171
• /
• pp.823-832
• /
• 2006

Precipitation is the most important component to the study of water and energy cycle in hydrology. In this study we investigate rainfall retrieval uncertainty from different sources of remotely sensed precipitation field and then probable error propagation in the simulation of hydrologic variables especially, runoff on different vegetation cover. Two remotely sensed rainfall retrievals (space-borne IR-only and ground radar rainfall) are explored and compared visually and statistically. Then, an offline Community Land Model (CLM) is forced with in situ meteorological data to simulate the amount of runoff and determine their impact on model predictions. A fundamental assumption made in this study is that CLM can adequately represent the physical land surface processes. Results show there are big differences between different sources of precipitation fields in terms of the magnitude and temporal variability. The study provides some intuitions on the uncertainty of hydrologic prediction via the interaction between the land surface and near atmosphere fluxes in the modelling approach. Eventually it will contribute to the understanding of water resources redistribution to the climate change in Korean Peninsula.

• Korean Journal of Remote Sensing
• /
• v.37 no.2
• /
• pp.321-335
• /
• 2021

Particulate matter (PM10 and PM2.5 with a diameter less than 10 and 2.5 ㎛, respectively) can be absorbed by the human body and adversely affect human health. Although most of the PM monitoring are based on ground-based observations, they are limited to point-based measurement sites, which leads to uncertainty in PM estimation for regions without observation sites. It is possible to overcome their spatial limitation by using satellite data. In this study, we developed machine learning-based retrieval algorithm for ground-level PM10 and PM2.5 concentrations using aerosol parameters from Geostationary Ocean Color Imager (GOCI) satellite and various meteorological parameters from a numerical weather prediction model during January to December of 2019. Gradient Boosted Regression Trees (GBRT) and Light Gradient Boosting Machine (LightGBM) were used to estimate PM concentrations. The model performances were examined for two types of feature sets-all input parameters (Feature set 1) and a subset of input parameters without meteorological and land-cover parameters (Feature set 2). Both models showed higher accuracy (about 10 % higher in R2) by using the Feature set 1 than the Feature set 2. The GBRT model using Feature set 1 was chosen as the final model for further analysis(PM10: R2 = 0.82, nRMSE = 34.9 %, PM2.5: R2 = 0.75, nRMSE = 35.6 %). The spatial distribution of the seasonal and annual-averaged PM concentrations was similar with in-situ observations, except for the northeastern part of China with bright surface reflectance. Their spatial distribution and seasonal changes were well matched with in-situ measurements.

• Journal of the Korea Computer Graphics Society
• /
• v.17 no.4
• /
• pp.11-21
• /
• 2011

This paper proposes a method for decomposing a Hangul glyph of outline fonts into its initial, medial and final components using statistical-structural information. In a font family, the positions of components are statistically consistent and the stroke relationships of a Hangul character reflect its structure. First, we create the component histograms that accumulate the shapes and positions of the same components. Second, we make pixel clusters from character image based on pixel direction probabilities and extract the candidate strokes using position, direction, size of clusters and adjacencies between clusters. Finally, we find the best structural match between candidate strokes and predefined character model by relaxation labeling. The proposed method in this paper can be used for a study on formative characteristics of Hangul font, and for a font classification/retrieval system.