• Journal of Internet Computing and Services
• /
• v.13 no.4
• /
• pp.1-10
• /
• 2012

In this paper, an optimal feature selection method considering sematic of features, which is preprocess of document classification is proposed. The feature selection is very important part on classification, which is composed of removing redundant features and selecting essential features. LSA (Latent Semantic Analysis) for considering meaning of the features is adopted. However, a supervised LSA which is suitable method for classification problems is used because the basic LSA is not specialized for feature selection. We also apply GA (Genetic Algorithm) to the features, which are obtained from supervised LSA to select better feature subset. Finally, we project documents onto new selected feature subset and classify them using specific classifier, SVM (Support Vector Machine). It is expected to get high performance and efficiency of classification by selecting optimal feature subset using the proposed hybrid method of supervised LSA and GA. Its efficiency is proved through experiments using internet news classification with low features.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2012.10a
• /
• pp.704-706
• /
• 2012

In this paper, we proposes a document summarization method using the term weighting. The proposed method can minimize the user intervention to use the pseudo relevance feedback. It also can improve the quality of document summaries because the inherent semantic of the sentence set are well reflected by term weighting derived from semantic feature.

• Journal of KIISE:Software and Applications
• /
• v.26 no.4
• /
• pp.584-594
• /
• 1999

의미 체계는 한국어 기초어휘에 대한 개념지식을 구축하는데 기본이 될 뿐만 아니라, 문장 분석시의 구조적 모호성과 단어 의미 모호성을 해소하는 중요한 단서를 제공할수 있다. 이러한 의미 체계가 실용적으로 여러 응용 시스템에서 사용되기 위해서는 광범위하고 타당한 자료를 바탕으로 하여 객관적인 방법으로 설정되어야한다. 국어 사전의 뜻풀이말에서의 상위개념을 표제어의 상위어로 선정하는 bottom-up 방식으로 구축하였던 한국어 명사의미체계는 사전편찬시의 비일관적인 뜻풀이말의 기술에 따른 여러 문제점이 있었다. 본 연구에서는 이러한 문제점들을 해결하기 위해서 사전 뜻풀이말에서 상위개념을 수식하는 어절과 용언의 의미호응관계에서 상위개념의 의미속성을 추출하고, 이들 의미 속성에 의한 명사의미체계를 구축하고 이를 바탕으로 명사의미 TAG를 설정할수 있다.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2013.10a
• /
• pp.305-306
• /
• 2013

In this paper, we proposes a document summarization method using the weighting based on cloud. The proposed method can minimize the user intervention to use the relevance feedback. It also can improve the quality of document summaries because the inherent semantic of the sentence set are well reflected by term weighting derived from semantic feature using nonnegative matrix factorizaitno based cloud.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2013.10a
• /
• pp.968-969
• /
• 2013

In this paper, we proposes a document summarization method using the weighting based on cloud. The proposed method can minimize the user intervention to use the relevance feedback. It also can improve the quality of document summaries because the inherent semantic of the sentence set are well reflected by term weighting derived from semantic feature using nonnegative matrix factorizaitno based cloud.

• Korean Journal of Remote Sensing
• /
• v.35 no.2
• /
• pp.279-288
• /
• 2019

We evaluated the land cover classification performance of SegNet, which features semantic segmentation of aerial imagery. We selected four semantic classes, i.e., urban, farmland, forest, and water areas, and created 2,000 datasets using aerial images and land cover maps. The datasets were divided at a 8:2 ratio into training (1,600) and validation datasets (400); we evaluated validation accuracy after tuning the hyperparameters. SegNet performance was optimal at a batch size of five with 100,000 iterations. When 200 test datasets were subjected to semantic segmentation using the trained SegNet model, the accuracies were farmland 87.89%, forest 87.18%, water 83.66%, and urban regions 82.67%; the overall accuracy was 85.48%. Thus, deep learning-based semantic segmentation can be used to classify land cover.

• Journal of the Korean Institute of Traditional Landscape Architecture
• /
• v.35 no.1
• /
• pp.36-47
• /
• 2017

In this research, sematic comparative study on formative idea and landscape elements composition was made between Soswaewon which was a remote villa garden in Joseon Dynasty and Canglang Pavilion which was built at a time when Neo-confucianism had been created in Song dynasty. From which the summary was resulted as below: Firstly, Soswaewon was fully devoted to Neo-confucianism among Korean remote villa gardens, Canglang-Pavilion was deeply affiliated with Neo-confucianism the deep-rooted thought. Adding a comment, as a richly historied Chinese private garden it was the best preserved even now and the farmost point from now and approximately 500 years advanced than Soswaewon, moreover Sosunheum the builder belonged to gentry society in Song dynasty. Secondly, Both Soswaewon and Canglang-Pavilion were remarkable thank to nature friendly concept which was deeply rooted in the eremitic thought(隱逸思想). Most of all, it might be interpreted the location of Soswaewon lay down in a mountain, as to be swallow hidden(小隱) but that of Canglang-Pavilion lay down in a city, as to be middle hidden(中隱). Thirdly, approaching with comparative study to gardens' factors implicated in symbolization when naming both Soswaewon and Canglang-Pavilion, neo-confucianism in Song dynasty in which diverse thoughts ranging Confucianism and Buddhism and Taoism had formed ensemble and developed was deeply built in Soswaewon, as to be in consistency to nature and to be a basis for comprehending diverse symbolization factors found in Canglang-Pavilion. Fourthly, given that forms or factors of gardens was tangibly came up with thoughts' variation we surely underwent to comparative study. Through which we could know Soswaewon was made by maximizing use of nature but minimizing addition of artificial things. In other words, it was trial of expressing semantic feature of the site in a way blending natural circumstance and liberal one. One of the representatives, Daebongdae in Soswaewon was a site where the owner's idea came up with. Scenary was recreated and nature was represented in interspace of Canglang-Pavilion with the high wall around the border, for which it renounced the world. To the end, it was understood there was the location characteristic of Canglang-Pavilion which let us look down from mountain and take some time for introspection. It might be said that the cultural root was in common between Korea and China, however it was found there were lots of differences in forms and features of gardens. We were able to interpret that social and cultural background were led to gardens' formation in which individual characteristic of two nations were blended, from which difference was resulted.