• 한국멀티미디어학회논문지
• /
• 제21권12호
• /
• pp.1425-1430
• /
• 2018

In this paper, we propose a CNN based deep learning algorithm for semantic segmentation of images. In order to improve the accuracy of semantic segmentation, we combined pixel level object classification and image level object classification. The image level object classification is used to accurately detect the characteristics of an image, and the pixel level object classification is used to indicate which object area is included in each pixel. The proposed network structure consists of three parts in total. A part for extracting the features of the image, a part for outputting the final result in the resolution size of the original image, and a part for performing the image level object classification. Loss functions exist for image level and pixel level classification, respectively. Image-level object classification uses KL-Divergence and pixel level object classification uses cross-entropy. In addition, it combines the layer of the resolution of the network extracting the features and the network of the resolution to secure the position information of the lost feature and the information of the boundary of the object due to the pooling operation.

• 한국의류학회지
• /
• 제19권6호
• /
• pp.857-865
• /
• 1995

The main purposes of this study were to find out the new classification system of the clothing evaluative criteria(CEC), 3nd to clear up the relationshiops of new classification system and the existing classification systems. For this purpose, the existing literatures related with the CEC(the classification system, and the variables) were investigated. The results of the study were as follows: 1. The existing classification systems were 'the intrinsic. non-intrinsic classification', 'the level classification', 'the purchase process classification' The new classification system of the CEC is based on 'the view-point of subjets'. The system was divided into the point of clothing itself, the wearer, the other, the wearing situation. The wearer's point of view is divided into the point of the value, and the physical characteristics of wearer 2. The image was included as the concept of the CEC, and the image classification could be suggested. 3. The relationships of the classification systems were as follows: $\circled1$ The intrinsic. non-intrinsic classification system included the level classification, the view-point classification, the image classification, and the buying process classificstion. $\circled2$ The level classification, the view.point classification, and the image classification were linked mutually, but the buying process classification is seperated from these classifications.

• Asia pacific journal of information systems
• /
• 제20권2호
• /
• pp.23-37
• /
• 2010

Conventional Support Vector Machines (SVMs) have been utilized as classifiers for binary classification problems. However, certain real world problems, including corporate bond rating, cannot be addressed by binary classifiers because these are multi-class problems. For this reason, numerous studies have attempted to transform the original SVM into a multiclass classifier. These studies, however, have only considered nominal classification problems. Thus, these approaches have been limited by the existence of multiclass classification problems where classes are not nominal but ordinal in real world, such as corporate bond rating and multiclass customer classification. In this study, we adopt a novel multiclass SVM which can address ordinal classification problems using ordinal pairwise partitioning (OPP). The proposed model in our study may use fewer classifiers, but it classifies more accurately because it considers the characteristics of the order of the classes. Although it can be applied to all kinds of ordinal multiclass classification problems, most prior studies have applied it to finance area like bond rating. Thus, this study applies it to a real world customer level classification case for implementing customer relationship management. The result shows that the ordinal multiclass SVM model may also be effective for customer level classification.

• 한국컴퓨터정보학회논문지
• /
• 제23권11호
• /
• pp.31-41
• /
• 2018

Text classification has been studied for a long time in the Natural Language Processing field. In this paper, we propose an article- and paragraph-level genre classification system using Word2Vec-based LSTM, GRU, and CNN models for large-scale English corpora. Both article- and paragraph-level classification performed best in accuracy with LSTM, which was followed by GRU and CNN in accuracy performance. Thus, it is to be confirmed that in evaluating the classification performance of LSTM, GRU, and CNN, the word sequential information for articles is better than the word feature extraction for paragraphs when the pre-trained Word2Vec-based word embeddings are used in both deep learning-based article- and paragraph-level classification tasks.

• Asia pacific journal of information systems
• /
• 제29권4호
• /
• pp.771-788
• /
• 2019

Deep learning techniques such as Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs) show superior performance in text classification than traditional approaches such as Support Vector Machines (SVMs) and Naïve Bayesian approaches. When using CNNs for text classification tasks, word embedding or character embedding is a step to transform words or characters to fixed size vectors before feeding them into convolutional layers. In this paper, we propose a parallel word-level and character-level embedding approach in CNNs for text classification. The proposed approach can capture word-level and character-level patterns concurrently in CNNs. To show the usefulness of proposed approach, we perform experiments with two English and three Korean text datasets. The experimental results show that character-level embedding works better in Korean and word-level embedding performs well in English. Also the experimental results reveal that the proposed approach provides better performance than traditional CNNs with word-level embedding or character-level embedding in both Korean and English documents. From more detail investigation, we find that the proposed approach tends to perform better when there is relatively small amount of data comparing to the traditional embedding approaches.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제8권7호
• /
• pp.2261-2280
• /
• 2014

Recently, network traffic has become more complex and diverse due to the emergence of new applications and services. Therefore, the importance of application-level traffic classification is increasing rapidly, and it has become a very popular research area. Although a lot of methods for traffic classification have been introduced in literature, they have some limitations to achieve an acceptable level of performance in real-time application-level traffic classification. In this paper, we propose a novel application-level traffic classification method using payload size sequence (PSS) signature. The proposed method generates unique PSS signatures for each application using packet order, direction and payload size of the first N packets in a flow, and uses them to classify application traffic. The evaluation shows that this method can classify application traffic easily and quickly with high accuracy rates, over 99.97%. Furthermore, the method can also classify application traffic that uses the same application protocol or is encrypted.

• Journal of Advanced Marine Engineering and Technology
• /
• 제28권4호
• /
• pp.611-617
• /
• 2004

Pixel classification is one of basic image processing issues. The general characteristics of the pixels belonging to various classes are discussed and the radical principles of pixel classification are given. At the same time. a pixel classification scheme based on image direction measure is proposed. As a typical application instance of pixel classification, an adaptive multi-level median filter is presented. An image can be classified into two types of areas by using the direction information measure, that is. smooth area and edge area. Single direction multi-level median filter is used in smooth area. and multi-direction multi-level median filter is taken in the other type of area. What's more. an adaptive mechanism is proposed to adjust the type of the filters and the size of filter window. As a result. we get a better trade-off between preserving details and noise filtering.

• 경영과학
• /
• 제10권1호
• /
• pp.1-22
• /
• 1993

Systemetic classification of a technology is critical to the development of technology strategy. This paper suggests a new technology classification scheme for computer and communication : a two-level scheme. Technology is first classified by its role and function in the upper level which forms a 2 * 2 matrix. The technology is then further classified into the lower level of 3 classes by associations among technology elements. Thus, a new classification scheme of 2 * 2 * 3 matrix is proposed for the computer and communication technology.

• 대한가정학회지
• /
• 제40권5호
• /
• pp.147-158
• /
• 2002

This study was intended to make an analysis of the classification of Internet consumer information provided on the Internet and its realities with a focus on the realities of information characteristics and technological level to provide systematized consumer information for consumers. As a result, the following findings were obtained: First, it was found that information on consumer panel research, anti-site and consumerism is lacking in terms of the type of information currently provided in the Internet consumer information classification. Second, it was found that up-to-datensess, communically and connectivity were high in the level of Internet consumer information characteristics. Third, it was found that publicity was realities well carried out in terms of the technological level of Internet consumer information but that the realities of publicity was still tacking. Fourth, it is thought that it is necessary to construct the Internet consumer information provision site through several eve-level programs for elementary, middle and high school students in a systematic fashion in order to provide diverse consumers with Internet consumer information at several levels.

• 한국환경복원기술학회지
• /
• 제2권1호
• /
• pp.29-37
• /
• 1999

Land cover classification is an essential basic information in natural environment management; however, land cover classification studies in Korea have not yet been proceeded to a sufficient level. At the present, only a limited number of the precedent studies that only cover definite city area has been conducted. Furthermore, there is almost no research conducted on the land cover classification schemes that could accurately classify the Korea's land cover conditions. This study primarily focuses on the land cover classification scheme which carries the most urgent priority in order to classify and to map out the Korean land cover conditions. In order to develop the most suitable land cover classification scheme, many foreign land cover classification cases and projects that are being carried out were reviewed in depth. The land cover classification scheme this study proposes comprises 3 levels : The first level consists of 7 different classes; the second level consists of 22 different classes; and the third level is made up of 50 classes. The land cover classification map will serve many important roles in natural environment management, such as the conjecture of natural habitats and estimation of oxygen production or carbon dioxide absorption capability of a forest. In water pollution modelling, the land cover classification data can be used to estimate and locate non-point sources of water pollution. If applied to a watershed, modelling it will allow to estimate the total amount of pollution from non-point sources of pollution in the water shed. The land cover classification data will also be good as a barometer data that determines defusion of air pollutants in air pollution modelling.