• Journal of KIISE
• /
• v.44 no.12
• /
• pp.1290-1295
• /
• 2017

Document classification, a task of classifying the category of each document based on text, is one of the fundamental areas for natural language processing. Document classification may be used in various fields such as topic classification and sentiment classification. Neural network models for document classification can be divided into two categories: word-level models and character-level models that treat words and characters as basic units respectively. In this study, we propose a neural network model that combines character-level and word-level models to improve performance of document classification. The proposed model extracts the feature vector of each word by combining information obtained from a word embedding matrix and information encoded by a character-level neural network. Based on feature vectors of words, the model classifies documents with a hierarchical structure wherein recurrent neural networks with attention mechanisms are used for both the word and the sentence levels. Experiments on real life datasets demonstrate effectiveness of our proposed model.

• Proceedings of the Korean Society of Crop Science Conference
• /
• 2000.11a
• /
• pp.188-190
• /
• 2000

• Proceedings of the Korean Society of Crop Science Conference
• /
• 2004.04a
• /
• pp.86-87
• /
• 2004

• Proceedings of the Korean Society of Crop Science Conference
• /
• 2004.04a
• /
• pp.88-89
• /
• 2004

• Communications of the Korean Mathematical Society
• /
• v.9 no.3
• /
• pp.521-529
• /
• 1994

This paper studies the realization of irreducible unitary representations of $GL_2(R)$ and $GL_2(C)$ by Bargmann's classification[1]. Since the representations of general matrix groups can be obtained by the extensions of characters of a special linear group, we shall follow to a large extent the pattern of the results in [5], [6], and [8]. This article is divided into two sections. In the first section we describe the realization of principal series and discrete series and complementary series of $GL_2(R)$. The last section is devoted to the derivation of principal series and complementary series of $GL_2(C).

• Journal of Plant Biology
• /
• v.37 no.2
• /
• pp.245-252
• /
• 1994

Pollens of 20 species of Sedum were investigated with a scanning electron microscope. The pollen morphology of Sedum was rather variable, within particular species or even within a single inflorescence. Differences occurred in the number and shape of apertures and surface sculpture. Besides 3-colporate, various aperture types including 2-syncolporate, 3-syncolporate, 40stephanocolporate, 5-stephanocolporate, zonate, and irregular types were found in a single specimen. Also, striate-rugulose and psilate sculpture were found in S. viviparum. No correlation was found between the pollen morphology and the floral formula. Pollen characters appeared to be not useful for infrageneric classification of Korean Sedum.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1995.10b
• /
• pp.218-223
• /
• 1995

A new design scheme of a binary decision tree is proposed. In this scheme a binary decision tree is constructed by using genetic algorithm and FCM algorithm. At each node optimal or near-optimal feature or feature subset among all the available features is selected based on fitness function in genetic algorithm which is inversely proportional to classification error, balance between cluster, number of feature used. The proposed design scheme is applied to the handwtitten alphabetic characters. Experimental results show the usefulness of the proposed scheme.

• Proceedings of the Korean Society of Crop Science Conference
• /
• 2000.04a
• /
• pp.110-111
• /
• 2000

• Proceedings of the Korean Society of Crop Science Conference
• /
• 2001.09a
• /
• pp.158-159
• /
• 2001

• Korean Journal of Medicinal Crop Science
• /
• v.16 no.4
• /
• pp.218-224
• /
• 2008

This study was conducted to obtain the basic data for using the Artemisia genetic resources as a medicinal crop. 24 taxa including Artemisia capillaris Thunb. were analyzed by principal component analysis of 25 characters and cluster analysis for classification. In Principal components analysis of individuals of taxa using 25 morphological characters of reproductive organ, the first, the second, the third and the fourth components contributed 44.73%, 16.86%, 8.88%, and 7.07% of the variations, respectively. The cumulative contribution from the first to the fourth principal components was 77.56%. In cluster analysis, taxa of Artemisia L. was seperated 3 group by 25 morphological characters of reproductive organ, but it didn't completely coincident with Kitamura classification.