• The Journal of Korean Obstetrics and Gynecology
• /
• v.31 no.1
• /
• pp.147-154
• /
• 2018

Objectives: This study aims to examine the improvement against facial wrinkles after needle-embedding therapy. Methods: The study participants were 10 women treated with needle-embedding therapy in the Korean Medical Clinic from September, 2011 to August, 2014. The subjects were treated every ten days for twenty days, totally three times, and the result of treatment was evaluated five times, such as pre-treatment, after each treatments, one month and two months later of the last treatment with ARAMO-SG. Results: 1. Depth and range of facial wrinkles, facial skin pore test, and sensitivity test were visibly improved after needle-embedding therapy and follow-up compared pre-treatment (p<0.05). 2. The significant improvements in facial skin texture test was not observed and sustained until the follow-up measurement at 2 months. 3. Any adverse reaction related to needle-embedding therapy did not happen. Conclusions: This study suggests that needle-embedding therapy can improve facial wrinkles. The positive results of this study support the requirement for additional research investigating the efficacy of needle-embedding therapy in women.

• 한국전자현미경학회:학술대회논문집
• /
• 1995.06a
• /
• pp.8-8
• /
• 1995

• Proceedings of the Korean Society for Technology of Plasticity Conference
• /
• 2003.10a
• /
• pp.241-244
• /
• 2003

The productivity of hydroforming process can be increased by combining pre-forming process and post-forming process such as the bending, piercing and the embedding process. Therefore in this study, integrated studies on the hydro-embedding technology have been performed by analyzing the deformed mode of the tubes and the optimal process parameters. In the case of the embedding test the characteristics of the embedded parts, such as the shape of the screw tip, screw thread and shape of thread were investigated at various process conditions. To measure the clamping force between the embedded part and the tube, special measuring device was used.

• Transactions of Materials Processing
• /
• v.14 no.9 s.81
• /
• pp.756-763
• /
• 2005

The applicability and productivity of hydroforming process can be increased by combining pre- and post-forming processes such as the bending, piercing and embedding process. For the fabrication of automotive parts, the hollow bodies with connecting nuts are widely used to connect parts together. Hollow body with connecting nuts has been conventionally fabricated by welding nuts or screwing in autobody screws. It requires multiple steps and devices fur the welding and/or screwing Therefore in this study, hydro-embedding process that combines the hydraulic embedding of connecting element(nut) with hydroforming process is investigated. Studies on the hydro-embedding technology have been performed to optimize the shape of the connecting element by analyzing the deformed mode of the embedded tube The effects of the shape of the screw tip, screw thread and shape of thread on the connection force between the tube and the connecting element have been investigated to optimize the shape of connecting element. Finite element analysis has also been performed to provide deformation behaviors of the tube surrounding a hole produced by hydro-embedding.

• Journal of KIISE
• /
• v.44 no.3
• /
• pp.306-313
• /
• 2017

Named entity recognition (NER) seeks to locate and classify named entities in text into pre-defined categories such as names of persons, organizations, locations, expressions of times, etc. Recently, many state-of-the-art NER systems have been implemented with bidirectional LSTM CRFs. Deep learning models based on long short-term memory (LSTM) generally depend on word representations as input. In this paper, we propose an approach to expand word representation by using pre-trained word embedding, part of speech (POS) tag embedding, syllable embedding and named entity dictionary feature vectors. Our experiments show that the proposed approach creates useful word representations as an input of bidirectional LSTM CRFs. Our final presentation shows its efficacy to be 8.05%p higher than baseline NERs with only the pre-trained word embedding vector.

• Proceedings of the Korean Society for Technology of Plasticity Conference
• /
• 2004.05a
• /
• pp.206-209
• /
• 2004

In the hydroforming process the number of process can be reduced by combining pre-forming process and post-forming process such as the bending, piercing and the embedding process. Integrated studies on the embedding manufacturing technology have been performed by analyzing the deformed mode of the tubes and the optimal process parameters. In this study, a simulation model that can prove clamping force between the clamping element and tube has been investigated by FEM. The characteristics of the embedded parts, such as the shape of the screw tip, screw thread and shape of thread were investigated at various clamping element conditions.

• Journal of the Korea Society of Computer and Information
• /
• v.25 no.5
• /
• pp.187-197
• /
• 2020

Recently, various methods of text embedding using deep learning algorithms have been proposed. Especially, the way of using pre-trained language model which uses tremendous amount of text data in training is mainly applied for embedding new text data. However, traditional pre-trained language model has some limitations that it is hard to understand unique context of new text data when the text has too many tokens. In this paper, we propose self-supervised learning-based fine tuning method for pre-trained language model to infer vectors of long-text. Also, we applied our method to news articles and classified them into categories and compared classification accuracy with traditional models. As a result, it was confirmed that the vector generated by the proposed model more accurately expresses the inherent characteristics of the document than the vectors generated by the traditional models.

• International Journal of Oral Biology
• /
• v.40 no.4
• /
• pp.183-187
• /
• 2015

The present study was aimed to evaluate the influence of glutaraldehyde (GA) concentration on multiple electron microscopic (EM) immunostaining using pre-embedding peroxidase and post-embedding immunogold method. Influence of various concentrations of GA included in the fixative on immuoreactivity was assessed in the multiple immunostaining using antisera against anti-transient receptor potential vanilloid 1 (TRPV1) for peroxidase staining and anti-GABA for immunogold labeling in the rat trigeminal caudal nucleus. Anti-TRPV1 antiserum had specificity in pre-embedding peroxidase staining when tissues were fixed with fixative containing paraformaldehyde (PFA) alone. Immunoreactivity for TRPV1 was specific in tissues fixed with fixative containing 0.5% GA at both perfusion and postfixation steps, though the immunoreactivity was weaker than in tissues fixed with fixative containing PFA alone. Tissues fixed with fixative containing 0.5% GA at the perfusion and postfixation steps showed specific immunogold staining for GABA. The results of the present study indicate that GA concentration is critical for immunoreactivity to antigens such as TRPV1 and GABA. This study also suggests that the appropriate GA concentration is 0.5% for multiple immunostaining with peroxidase labeling for TRPV1 and immunogold labeling for GABA.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.9
• /
• pp.35-46
• /
• 2023

In this study, we propose a Dual Approach methodology to enhance the accuracy of document classifiers by utilizing both contextual and keyword information. Firstly, contextual information is extracted using Google's BERT, a pre-trained language model known for its outstanding performance in various natural language understanding tasks. Specifically, we employ KoBERT, a pre-trained model on the Korean corpus, to extract contextual information in the form of the CLS token. Secondly, keyword information is generated for each document by encoding the set of keywords into a single vector using an Autoencoder. We applied the proposed approach to 40,130 documents related to healthcare and medicine from the National R&D Projects database of the National Science and Technology Information Service (NTIS). The experimental results demonstrate that the proposed methodology outperforms existing methods that rely solely on document or word information in terms of accuracy for document classification.

• International Journal of Contents
• /
• v.17 no.4
• /
• pp.1-15
• /
• 2021

Named entity recognition (NER) is a crucial task for NLP, which aims to extract information from texts. To build NER systems, deep learning (DL) models are learned with dictionary features by mapping each word in the dataset to dictionary features and generating a unique index. However, this technique might generate noisy labels, which pose significant challenges for the NER task. In this paper, we proposed DL-dictionary features, and evaluated them on two datasets, including the OntoNotes 5.0 dataset and our new infectious disease outbreak dataset named GFID. We used (1) a Bidirectional Long Short-Term Memory (BiLSTM) character and (2) pre-trained embedding to concatenate with (3) our proposed features, named the Convolutional Neural Network (CNN), BiLSTM, and self-attention dictionaries, respectively. The combined features (1-3) were fed through BiLSTM - Conditional Random Field (CRF) to predict named entity classes as outputs. We compared these outputs with other predictions of the BiLSTM character, pre-trained embedding, and dictionary features from previous research, which used the exact matching and partial matching dictionary technique. The findings showed that the model employing our dictionary features outperformed other models that used existing dictionary features. We also computed the F1 score with the GFID dataset to apply this technique to extract medical or healthcare information.