• International journal of advanced smart convergence
• /
• v.8 no.4
• /
• pp.75-81
• /
• 2019

Though recent advancement of deep learning methods have provided satisfactory results from large data domain, somehow yield poor performance on few-shot classification tasks. In order to train a model with strong performance, i.e. deep convolutional neural network, it depends heavily on huge dataset and the labeled classes of the dataset can be extremely humongous. The cost of human annotation and scarcity of the data among the classes have drastically limited the capability of current image classification model. On the contrary, humans are excellent in terms of learning or recognizing new unseen classes with merely small set of labeled examples. Few-shot learning aims to train a classification model with limited labeled samples to recognize new classes that have neverseen during training process. In this paper, we increase the backbone depth of the embedding network in orderto learn the variation between the intra-class. By increasing the network depth of the embedding module, we are able to achieve competitive performance due to the minimized intra-class variation.

• Journal of Korea Society of Digital Industry and Information Management
• /
• v.10 no.2
• /
• pp.203-211
• /
• 2014

In recent highly competitive business environment, collaboration has become one of the important business strategies for companies to survive and/or prosper. There are many different types of collaboration strategies, and it is crucial for companies to select the right ones according to the types of collaboration they require. To select the right type of collaboration options for business, in the past research, there have been two important criteria to classify collaboration types, namely governance (who makes key decisions - one kingpin participant or all players?) and membership (can anyone participate, or just select players?). In this research, we add a new classification criterion, recognition of collaboration purpose, which means whether collaborators know or do not know the purpose of collaboration in advance. Recently, we see many cases in which social media data are used in many unknown purposes a priori. In this research, we add such cases to develop new classification model.

• Journal of Korea Multimedia Society
• /
• v.23 no.2
• /
• pp.216-226
• /
• 2020

In this paper, a stacked convolutional autoencoder model is proposed in order to classify Alzheimer's disease with high accuracy in PET/CT images. The proposed model makes use of the latent space representation - which is also called the bottleneck, of the encoder-decoder architecture: The input image is sent through the pipeline and the encoder part, using stacked convolutional filters, extracts the most useful information. This information is in the bottleneck, which then uses Softmax classification operation to classify between Alzheimer's disease, Mild Cognitive Impairment, and Normal Control. Using the data from Dong-A University, the model performs classification in detecting Alzheimer's disease up to 98.54% accuracy.

• Journal of Information Technology Services
• /
• v.11 no.sup
• /
• pp.139-152
• /
• 2012

In the medical web forum, people share medical experience and information as patients and patents' families. Some people search medical information written in non-expert language and some people offer words of comport to who are suffering from diseases. Medical web forums play a role of the informative support and the emotional support. We propose the automatic classification model of articles in the medical web forum into the information support and emotional support. We extract text features of articles in web forum using text mining techniques from the perspective of linguistics and then perform supervised learning to classify texts into the information support and the emotional support types. We adopt the Support Vector Machine (SVM), Naive-Bayesian, decision tree for automatic classification. We apply the proposed model to the HealthBoards forum, which is also one of the largest and most dynamic medical web forum.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2012.10a
• /
• pp.153-158
• /
• 2012

Noise maps are produced according to 'The Method of making a Noise Map' in order to noise control efficiently, and prediction model to predict road traffic noise which may apply to Korean situation, include CRTN, RLS 90, NMPB, Nord 2000 and ASJ 2003. Of them, ASJ 2003, Japan's prediction model has not been verified for the application to Korean situation according to the classification of vehicle. In addition, ASJ 2003 was revised to ASJ 2008 recently, a classification for motorcycle was added. This study attempts to check the classification of vehicle in ASJ 2008 and 'The Method of making a Noise Map' to confirm the suitability of the application of them to Korean situation.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• v.9 no.2
• /
• pp.383-386
• /
• 2005

This study proposes a case based reasoning system with two dimensional reduction techniques. In this study, vertical and horizontal dimensions of the research data are reduced through hybrid feature and instance selection process using genetic algorithms. We applied the proposed model to customer classification model which utilizes customers' demographic characteristics as inputs to predict their buying behavior for the specific product. Experimental results show that the proposed technique may improve the classification accuracy and outperform various optimized models of typical CBR system.

• Journal of Korea Multimedia Society
• /
• v.25 no.6
• /
• pp.769-782
• /
• 2022

A study on agricultural automation became more important. In Korea, sweet persimmon farmers spend a lot of time and effort on classifying profitable persimmons. In this paper, we propose and implement an efficient grading algorithm for persimmons before shipment. We gathered more than 1,750 images of persimmons, and the images were graded and labeled for classifications purpose. Our main algorithm is based on EfficientDet object detection model but we implemented more exquisite method for better classification performance. In order to improve the precision of classification, we adopted a machine learning algorithm, which was proposed by PyCaret machine learning workflow generation library. Finally we acquired an improved classification model with the accuracy score of 81%.

• Journal of Korea Multimedia Society
• /
• v.25 no.10
• /
• pp.1524-1534
• /
• 2022

This study proposes the classification model of Minhwa genre based on object detection of deep learning. To detect unique Korean traditional objects in Minhwa, we construct custom datasets by labeling images using object keywords in Minhwa DB. We train YOLOv5 models with custom datasets, and classify images using predicted object labels result, the output of model training. The algorithm consists of two classification steps: 1) according to the painting technique and 2) genre of Minhwa. Through classifying paintings using this algorithm on the Internet, it is expected that the correct information of Minhwa can be built and provided to users forward.

• The Journal of the Acoustical Society of Korea
• /
• v.39 no.6
• /
• pp.615-621
• /
• 2020

In this paper, we propose a multi-site based earthquake event classification method using graph convolution networks. In the traditional earthquake event classification methods using deep learning, they used single-site observation to estimate seismic event class. However, to achieve robust and accurate earthquake event classification on the seismic observation network, the method using the information from the multi-site observations is needed, instead of using only single-site data. Firstly, our proposed model employs convolution neural networks to extract informative embedding features from the single-site observation. Secondly, graph convolution networks are used to integrate the features from several stations. To evaluate our model, we explore the model structure and the number of stations for ablation study. Finally, our multi-site based model outperforms up to 10 % accuracy and event recall rate compared to single-site based model.

• Journal of the Korea Safety Management & Science
• /
• v.20 no.1
• /
• pp.41-55
• /
• 2018

For several decades, attribute classification methods using the asymmetrical relationship between an attribute performance and the satisfaction of that attribute have been explored by numerous researchers. In particular, the Kano model, which classifies quality attributes into 5 elements using simple questionnaire and two-dimensional evaluation table, has gained popularity: Attractive, One-dimensional, Must-be, Indifferent, and Reverse quality. As Kano's model is well accepted, many literatures have introduced categorization methods using the Kano's evaluation table at attribute level. However, they applied different terminologies and classification criteria and this causes confusion and misunderstanding. Therefore, a criterion for quality classification at attribute level is necessary. This study is aimed to suggest a new attribute classification method that sub-categorizes quality attributes using 5-point ordinal point and Kano's two-dimensional evaluation table through an extensive literature review. For this, the current study examines the intrinsic and extrinsic problems of the well-recognized Kano model that have been used for measuring customer satisfaction of products and services. For empirical study, the author conducted a comparative study between the results of Kano's model and the proposed method for an e-learning case (33 attributes). Results show that the proposed method is better in terms of ease of use and understanding of kano's results and this result will contribute to the further development of the attractive quality theory that enables to understand both the customers explicit and implicit needs.