• The Journal of Society for e-Business Studies
• /
• v.26 no.1
• /
• pp.29-41
• /
• 2021

Recently, there are increasing attempts to utilize deep learning methodology in the fashion industry. Accordingly, research dealing with various fashion-related problems have been proposed, and superior performances have been achieved. However, the studies for fashion style classification have not reflected the characteristics of the fashion style that one outfit can include multiple styles simultaneously. Therefore, we aim to solve the multi-label classification problem by utilizing the dependencies between the styles. A multi-label recognition model based on a graph convolution network is applied to detect and explore fashion styles' dependencies. Furthermore, we accelerate model training and improve the model's performance through transfer learning. The proposed model was verified by a dataset collected from social network services and outperformed baselines.

• Journal of Korea Entertainment Industry Association
• /
• v.15 no.7
• /
• pp.249-257
• /
• 2021

Recently, interest and demand for facial information processing technology that can provide non-face-to-face identity authentication and access management service using smart devices, which is an essential environmental improvement for multi-use facilities, is increasing as a way to prevent the spread of infectious diseases worldwide and to cope with social measures. This study defines a multi-use facility classification system and applied service field to establish a continuous access control system, and measures to improve the usability of the service platform considering scalability through a dual access control system and personal/measurement information type analysis, and accordingly We would like to propose a service roadmap. In addition, it aims to improve the physical access management system service platform, which is a multi-use facility application service that requires one-time and multiple-use authentication according to usage. It is expected that the methodology of this study can be applied as a service platform of a logical access control system type in the future.

• Journal of Intelligence and Information Systems
• /
• v.22 no.3
• /
• pp.91-111
• /
• 2016

It is crucial for the social media based marketing practices to perform sentiment analyze the unstructured data written by the potential consumers of their products and services. In particular, when it comes to the companies which are interested in global business, the companies must collect and analyze the data from the social media of multinational settings (e.g. Youtube, Instagram, etc.). In this case, since the texts are multilingual, they usually translate the sentences into a certain target language before conducting sentiment analysis. However, due to the lack of cultural differences and highly qualified data dictionary, translated sentences suffer from misunderstanding the true meaning. These result in decreasing the quality of sentiment analysis. Hence, this study aims to propose a method to perform a multilingual sentiment analysis, focusing on Korean-Chinese cases, while avoiding language translations. To show the feasibility of the idea proposed in this paper, we compare the performance of the proposed method with those of the legacy methods which adopt language translators. The results suggest that our method outperforms in terms of RMSE, and can be applied by the global business institutions.

• Journal of the Korean Geographical Society
• /
• v.49 no.2
• /
• pp.113-138
• /
• 2014

Despite the growing numbers of regional problems (e.g. conflicts between the state and localities, inter-local conflicts, etc.) associated with the state-led developmental projects, the Korean social sciences have been unable to offer satisfying explanations and solutions to the regional problems. This is mainly because the existing works, which have been taken captured by the assumptions of "methodological nationalism", significantly lack the socio-spatial understandings of the state actions and the relations between the state and localities, thereby seeing the issues of regional development mainly in terms of either the economic efficiency defined at the national scale, or the plan rationality of the national bureaucrats. With this problem orientation in mind, this paper aims to explore the ways in which the state and localities are interacting, conflicting and negotiating with one another through the mediation of the state-led developmental projects. Focusing on the developmental processes of Masan Export Processing Zone from the mid-1960s to the early 1970s, it examines the multi-scalar processes through which the state-led industrial complex developmental processes have been influenced by the complex and dynamic interactions among social forces and actors acting at diverse geographical scales (e.g. the global, national, local, urban, etc.). This analysis shows that the regional policies of the Korean developmental state were more heavily influenced by the interactions, contestations, and collaborations among social forces and actors, acting in and through the state, at various geographical scales, rather than by the economic and techno-bureaucratic rationality.

• Korean Journal of Remote Sensing
• /
• v.33 no.5_2
• /
• pp.689-698
• /
• 2017

In this paper, a new classification method based on the combination of semi-supervised learning with spatial similarity of adjacent pixels is presented for crop classification in inaccessible areas. Iterative classification based on semi-supervised learning is applied to extract reliable training data from both the initial classification result with a small number of training data, and classification results of adjacent pixels are also considered to extract new training pixels with less uncertainty. To evaluate the applicability of the proposed method, a case study of the classification of field crops was carried out using multi-temporal Landsat-8 OLI acquired in the Daehongdan region, North Korea. From a case study, the misclassification of crops and forests, and isolated pixels in the initial classification result were greatly reduced by applying the proposed semi-supervised learning method. In addition, the combination of classification results of adjacent pixels for the extraction of new training data led to the great reduction of both misclassification results and isolated pixels, compared to the initial classification and traditional semi-supervised learning results. Therefore, it is expected that the proposed method would be effectively applied to classify areas in which it is difficult to collect sufficient training data.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.21 no.5
• /
• pp.22-27
• /
• 2020

In this paper, we study the methodology to improve productivity and transportation efficiency simultaneously in the manufacturing environment of injection plants which has multiple injection machines arranged in parallel. In general, the post-processes such as finishing are continuously arranged in the injection machine located in the lower level of the injection plants, and one or two workers in charge of post-processing are always arranged. Therefore injection plants have low productivity due to post-processing and the front of the injection machine is very crowded due to various logistics flows. In this paper, we propose the designing methodology of conveyor structure for integrating the post-processes arranged at each injection machine and transporting the injection products to the integrated post-process automatically. Specifically, we propose the models for computing the number of conveyor units into the integrated processes, and for finding the optimal combinations to connect each machines and the conveyors. The proposed model is for the total productivity improvement, which are productivity and transportation efficiency. By applying the proposed model to companies that produce injection parts used for the home appliances, we verify the applicability and the effect of improving productivity and transportation efficiency, which more than 40%.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.20 no.3
• /
• pp.293-301
• /
• 2007

We present a new global-local analysis with the aid of MLS(Moving Least Square) variable-node finite elements which can possess an arbitrary number of nodes on element master domain. It enables us to connect one finite element with a few finite elements without complex remeshing. Compared to other type global-local analysis, it does not require any superimposed mesh or need not solve the equilibrium equation twice. To demonstrate the performance of the proposed scheme, we will show several examples in relation to capturing highly local stress field using global-local analysis.

• Journal of the Korean School Mathematics Society
• /
• v.19 no.2
• /
• pp.123-151
• /
• 2016

The purpose of this study is to analyze the existing studies about the effect on private tutoring and EBS broadcasting which is considered as the alternative which was presented by education authorities. And this study attempts to measure the degree of the effects of private tutoring and EBS engagement on mathematics achievement using various quantitative methodologies. T-test and OLS multiple regression indicates some selection-bias which has positive direction compared to the PSM method closed to experimental design. So, it will be required to consider the methodology which measures the exact effect of institution or participation when we conduct the observational studies.

• Journal of Intelligence and Information Systems
• /
• v.13 no.1
• /
• pp.29-47
• /
• 2007

Acquiring context data in a timely and correct way is now regarded as one of the crucial characteristics of the proactive service which runs on ubiquitous computing environment. Moreover, context model should be well designed to provide a solid context-aware system. Since the ubiquitous computing systems aim to provide context-aware services everywhere with any available devices, legacy services which uses context models assuming single or limited domain should be extended enough to be useful even for multi-domain muli-services. This leads us to a motivation to build a generic context model with an appropriate type of model. Hence, the purpose of this paper is to propose a generic context model by assessing a variety of model types with a sort of evaluation measures.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.05a
• /
• pp.154-157
• /
• 2022

In the existing Tensorflow CNN environment, the object detection method is a method of performing object labeling and detection by Tensorflow itself. However, with the advent of YOLO, the efficiency of image object detection has increased. As a result, more deep layers can be built than existing neural networks, and the image object recognition rate can be increased. Therefore, in this paper, the detection ability and speed were compared and analyzed by designing an object detection system based on Darknet and YOLO and performing multi-layer construction and learning based on the existing convolutional neural network. For this reason, in this paper, a neural network methodology that efficiently uses Darknet's learning is presented.