• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.8
• /
• pp.3168-3186
• /
• 2020

Automatic topic essay generation (TEG) is a controllable text generation task that aims to generate informative, diverse, and topic-consistent essays based on multiple topics. To make the generated essays of high quality, a reasonable method should consider both diversity and topic-consistency. Another essential issue is the intrinsic link of the topics, which contributes to making the essays closely surround the semantics of provided topics. However, it remains challenging for TEG to fill the semantic gap between source topic words and target output, and a more powerful model is needed to capture the semantics of given topics. To this end, we propose a pretraining-based contextual self-attention (PC-SAN) model that is built upon the seq2seq framework. For the encoder of our model, we employ a dynamic weight sum of layers from BERT to fully utilize the semantics of topics, which is of great help to fill the gap and improve the quality of the generated essays. In the decoding phase, we also transform the target-side contextual history information into the query layers to alleviate the lack of context in typical self-attention networks (SANs). Experimental results on large-scale paragraph-level Chinese corpora verify that our model is capable of generating diverse, topic-consistent text and essentially makes improvements as compare to strong baselines. Furthermore, extensive analysis validates the effectiveness of contextual embeddings from BERT and contextual history information in SANs.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2003.10a
• /
• pp.892-896
• /
• 2003

XML is used in various fields such as interface format for data swapping between application between several various system passing over thing to mark to web browser simply. Accordingly, a lot of studies about system that can manage effectively and search XML document with formation of information, reusability, disposal and durability, portability are proceeding. In this paper, explain about XQL and document structure processor and language processor of quality and make contents of XML document by tree structure, structure information presents method that find XML document tree structure information that is correct on question using XQL while do parsing. Through this, described for design and embodiment of efficient XML document search system that use XQL that compose structure information of document in tree structure and is proposed in language of quality after do parsing absorbing XML document that is scattered on web.

• Journal of the Korean Geographical Society
• /
• v.35 no.1
• /
• pp.3-15
• /
• 2000

산업화와 더불어 심각해지고 있는 수질오염 문제를 해결하기 위해서는 여러 가지 수질관리 방안이 요구된다. 수질오염이 과거에는 국지적이었으나 점차 광범한 지역으로 확장됨에 다라 지속적인 수질 모니터링에 어려움이 따른다. 본 연구에서는 위성영상을 사용한 원격탐사 기법으로 수역의 수질환경 인자를 추출하고자 한다. 사용된 영상은 Landasat TM이며, 연구지역은 한강하류 지역이다. 수질분석 인자는 클로로필-a, 부유물질, 투명도 등을 선정하였으며, 수면분광반사율의 특징 및 수질인자별 처리기법을 개발하는데 목적을 두었다. 분광특성 분석결과를 요약하면, 첫 번째 스펙트럼 반사율 분석결과 클로로필-a의 농도는 0.4~0.5$\mu\textrm{m}$ 파장대역에서 낮은 반사치 경향을 보이며, 녹색파장대인 0.57$\mu\textrm{m}$ 부근에서 반사율이 높아진다. 두 번째 부유물질의 반사도는 농도가 증가할수록 0.8$\mu\textrm{m}$ 부근에서 상대적으로 낮은 반사율이 나타난다. 마지막으로 투명도가 낮은 수면은 0.55$\mu\textrm{m}$에서 높은 반사율 경향을 보인다. Landsat TM영상을 이용하여 주성분분석 및 비연산처리를 실시하여 수질분석을 시도한 결과를 보면 클로로필-a와 투명도는 제1주성분 영상 및 제2주성분 영상에서 현장 실측자료와 유사한 결과를 얻을 수 있었으며, 부유물질은 밴드 2와 밴드 4의 비연산처리를 통하여 분포도를 작성할 수 있었다. 이상의 결과들은 계절적 및 시간적 변화에 따라 파장대역이 달라질 수 있다. 그러므로 위성자료를 이용하여 보다 정확한 수질환경 인자를 추출하기 위해서는 현장실측 및 수역의 분광반사 특성을 지속적으로 조사하여야 한다.때문으로 경주 산사태와 포함-구릉포간 국도면의 산사태가 이 종류의 산사태에 속한다.열 인식의 신뢰도를 향상시킬수 있는 방법을 제안하였다.작성하여 최신 의료영상 처리 기법을 쉽게 임상에 적용하고 실험할 수 있는 장점이 있다. 지대에서 가능하였고, 파종기는 중생종보다 이르게 나타났다. 등숙만한출수기 기준의 안전작기는 조생종과 중생종은 태백고냉지대와 태백준고냉지대, 소백산간지대 일부지역을 제외한 다른 지역에서 설정되었고, 중만생종은 태백고냉지대, 태백준고냉지대, 동해안북부지대, 소백산간지대, 노령소백산간지대의 일부 지역은 벼 담수직파가 불가능하게 판단되었다. information on the regular basis of time and provide it when the users query over the Web-database gateway. The other approach is a shopping agent mechanism, which stores information on "how to shop" and the shopping agent collects the information of product items just after users query about the product and provide the information in real time or notify them by alerting service. Thirty nine shopping information services are compared and classified in this paper and they are extracted from "Naver" and "Yahoo! Korea". The final result shows that most services are just a

• Proceedings of the Korean Society for Bioinformatics Conference
• /
• 2003.10a
• /
• pp.66-72
• /
• 2003

Toxicogenomics is now emerging as one of the most important genomics application because the toxicity test based on gene expression profiles is expected more precise and efficient than current histopathological approach in pre-clinical phase. One of the challenging points in Toxicogenomics is the construction of intelligent database management system which can deal with very heterogeneous and complex data from many different experimental and information sources. Here we present a new Toxicogenomics database developed as a part of 'Toxicogenomics for Efficient Safety Test (TEST) project'. The TEST database is especially focused on the connectivity of heterogeneous data and intelligent query system which enables users to get inspiration from the complex data sets. The database deals with four kinds of information; compound information, histopathological information, gene expression information, and annotation information. Currently, TEST database has Toxicogenomics information fer 12 molecules with 4 efficacy classes; anti cancer, antibiotic, hypotension, and gastric ulcer. Users can easily access all kinds of detailed information about there compounds and simultaneously, users can also check the confidence of retrieved information by browsing the quality of experimental data and toxicity grade of gene generated from our toxicology annotation system. Intelligent query system is designed for multiple comparisons of experimental data because the comparison of experimental data according to histopathological toxicity, compounds, efficacy, and individual variation is crucial to find common genetic characteristics .Our presented system can be a good information source for the study of toxicology mechanism in the genome-wide level and also can be utilized fur the design of toxicity test chip.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.16 no.2
• /
• pp.101-113
• /
• 2013

The concern on environmental quality makes the landscape analysis more important than before ever. For the landscape analysis, selection of viewpoint is one of most important stage. Because of its subjectiveness, the conventional viewpoint selection method often missed some viewpoints of importance. The purpose of this study is to develop a viewpoint selection method for landscape analysis using GIS data and techniques. During the viewpoint selection process, spatial and attribute data from several GIS systems were hired. Query and overlay methods were mainly adapted for analysis to find out meaningful viewpoints. The 3D simulation analysis on DEM(Digital Elevation Model) was used for every selected viewpoint to examine wether the view target is screened out or not. Application study at a sample site showed some omissions of good viewpoints without any screening. It also exhibited the possibility to reduce time and cost for the viewpoint selection process of landscape analysis. For the progress of applicability, GIS data analysis process have to be improved and more modules such as automatic screening analysis system on selected viewpoint have to be developed.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.27 no.6C
• /
• pp.544-561
• /
• 2002

In this paper, we propose a new hybrid type of the routing protocol (Virtual Cluster-based Routing Protocol: VCRP) for mobile ad-hoc networks, based on a virtual cluster, which is defined as a narrow-sense network to exchange the basic information related to the routing among the adjacent nodes. This particular approach combines advantage of proactive routing protocol (PRP), which immediately provides the route collecting the network-wide topological and metric information, with that of reactive routing protocol, which relies on the route query packet to collect the route information on its way to the destination without exchanging any information between nodes. Furthermore, it also provides the back-up route as a byproduct, along with the optimal route, which leads to the VCBRP (Virtual Cluster-based Routing Protocol with Backup Route) establishing the alternative route immediately after a network topology is changed due to degradation of link quality and terminal mobility, Our simulation studies have shown that the proposed routing protocols are robust against dynamics of network topology while improving the performances of packet transfer delay, link failure ratio, and throughput over those of the existing routing protocols without much compromising the control overhead efficiency.

• KIPS Transactions on Software and Data Engineering
• /
• v.7 no.9
• /
• pp.351-360
• /
• 2018

The LWR (Locally Weighted Regression) model, which is traditionally a lazy learning model, is designed to obtain the solution of the prediction according to the input variable, the query point, and it is a kind of the regression equation in the short interval obtained as a result of the learning that gives a higher weight value closer to the query point. We study on an incremental ensemble learning approach for LWR, a form of lazy learning and memory-based learning. The proposed incremental ensemble learning method of LWR is to sequentially generate and integrate LWR models over time using a genetic algorithm to obtain a solution of a specific query point. The weaknesses of existing LWR models are that multiple LWR models can be generated based on the indicator function and data sample selection, and the quality of the predictions can also vary depending on this model. However, no research has been conducted to solve the problem of selection or combination of multiple LWR models. In this study, after generating the initial LWR model according to the indicator function and the sample data set, we iterate evolution learning process to obtain the proper indicator function and assess the LWR models applied to the other sample data sets to overcome the data set bias. We adopt Eager learning method to generate and store LWR model gradually when data is generated for all sections. In order to obtain a prediction solution at a specific point in time, an LWR model is generated based on newly generated data within a predetermined interval and then combined with existing LWR models in a section using a genetic algorithm. The proposed method shows better results than the method of selecting multiple LWR models using the simple average method. The results of this study are compared with the predicted results using multiple regression analysis by applying the real data such as the amount of traffic per hour in a specific area and hourly sales of a resting place of the highway, etc.

• Journal of Society of Preventive Korean Medicine
• /
• v.20 no.1
• /
• pp.111-124
• /
• 2016

Objectives : Treatment method using drugs has already been used in Korean medicine for a long time. Moreover, database has been developed and utilized for more efficient management of the treatments that use drugs. Most of such database related to knowledge on drugs is composed of origin, efficacy, temperament, ingredients and examples of application of the standardized drugs. Communication with knowledge information in other specialized areas is also accomplished by using the efficacies and ingredients with the drugs. In this study, we aimed to make data structure of the terminologies that represent the manufacturing process of herbs. However, in spite of the fact that the manufacturing process of the drugs imparts effect on their efficacies and ingredients, details of the manufacturing processes are quite limited to simple text sentences, thereby resulting in substantially lower level of utilization and difficulties in systematic researches on various factors included in the manufacturing processes in comparison to other knowledge on drugs. Methods : This Study extracted the factors necessary in the development of database by executing morphological analysis of the manufacturing process of herbs. Results : The factors are 'Order', 'Act', 'Raw material', 'Tools', 'Supporting materials', 'Intensity', 'Duration Time', 'Interval', 'Focus', 'Repetition Number', 'Untill'. We were able to tell the difference of the manufacturing process with a simple structured query language and the factors. Conclusions : Morphological analysis of medicinal herbs manufacturing Process contributes to standardization with information of the manufacturing process. And it helps to creates a quality management system through the Database.

• Journal of Korea Spatial Information System Society
• /
• v.6 no.1 s.11
• /
• pp.3-17
• /
• 2004

With the rapid growth of the Internet geographic information services through the WWW such as a location-based service and so on. Web GISs (Geographic Information Systems) have also come to be a cluster-based architecture like most other information systems. That is, in order to guarntee high quality of geographic information service without regard to the rapid growth of the number of users, web GISs need cluster-based architecture that will be cost-effective and have high availability and scalability. This paper proposes the design of the cluster-based web GIS with high availability and scalability. For this, each node within a cluster-based web GIS consists of main memory spatial databases which accomplish role of caching by using data declustering and the locality of spatial query. Not only simple region queries but also the proposed system processed spatial join queries effectively. Compare to the existing method. Parallel R-tree spatial join for a shared-Nothing architecture, the result of simulation experiments represents that the proposed spatial join method achieves improvement of performance respectively 23% and 30% as data quantity and nodes of cluster become large.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.45 no.3
• /
• pp.58-67
• /
• 2008

In this paper, we first point out difficulties faced by CG artists in the shading process: (1) a lot of technical details on shaders required, (2) long rendering time, and (3) repeated trials-and-errors. To make them overcome such difficulties, we propose Shader Space Navigator, a system that efficiently searches for shaders similar to a given query shader from a shader database containing a large number of quality shaders. With Shader Space Navigator, CG artists find appropriate shaders from the database that are very close to the final result shader, and thus complete the shading process easily by slightly tuning some attributes of those shaders. Thus, the CG artists can create their final shaders in an intuitive and efficient way without a large number of time-consuming rendering processes. Also, we deal with implementation issues related to Shader Space Navigator and constructing an abundant shader database in detail.