• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.05a
• /
• pp.108-111
• /
• 2022

Applying deep learning to machine vision systems for defect detection of products requires vast amounts of training data about various defect cases. However, since data imbalance occurs according to the type of defect in the actual manufacturing industry, it takes a lot of time to collect product images enough to generalize defect cases. In this paper, we apply a Siamese neural network that can be learned with even a small amount of data to product defect detection, and modify the image pairing method and contrastive loss function by properties the situation of product defect image data. We indirectly evaluated the embedding performance of Siamese neural networks using AUC-ROC, and it showed good performance when the images only paired among same products, not paired among defective products, and learned with exponential contrastive loss.

• English Language & Literature Teaching
• /
• v.13 no.2
• /
• pp.155-176
• /
• 2007

An attempt to investigate the effect of gender differences on Korean students' EFL learning orientation was made. To explain a Korean EFL learning context, three criteria (cultural distance between the target country and the host country, communicative needs of the TL, the status of the TL in the host country) are adopted. Moreover, as a contrastive FL learning context from the Imposed FL learning context, a FL learning context where there is a substantial cultural distance from the TL community, communication needs of the TL do not exist, and the TL enjoys a special educational and socioeconomic status in the host country, a concept of an Integrative FL learning context is newly brought up in this paper. As the result of a questionnaire conducted in four different high schools, female learners can be claimed to be more internalized about academic and socioeconomic benefits the TL entails for their social advancement and overcoming inequality between men and women in society, albeit insignificant numerical data.

• East Asian mathematical journal
• /
• v.30 no.2
• /
• pp.123-148
• /
• 2014

The purpose of this article described the effects of mentoring program to the underachiever's attitude in learning Mathematics. In order to do this research, a mentoring class had been carried out for 30 weeks, with two students who was underachievers in mathematics. It was carried out contrastive analysis on the student's learning attitude in the math class before and after the mentoring, using a questionnaire of learning attitude in math with 40 questions, filling the blanks in sentence, recording files of class, and mentoring journals. As a result, before taking the mentoring class, students who were participated in the mentoring program were negative and low in five subordinate concepts(learning attitude, such as their tendency, interest, desire, confidence, attitude, and studying habits in math). However, after the mentoring class, the two students were remarkably changed in positive way on the five concepts in learning mathematics. Therefore, it is helpful to underachievers in mathematics if the mentoring program is used since it yields a positive impact on the learning attitude.

• International journal of advanced smart convergence
• /
• v.5 no.3
• /
• pp.8-15
• /
• 2016

In recent years, one of deep learning models called Deep Belief Network (DBN) which formed by stacking restricted Boltzman machine in a greedy fashion has beed widely used for classification and recognition. With an ability to extracting features of high-level abstraction and deal with higher dimensional data structure, this model has ouperformed outstanding result on image and speech recognition. In this research, we assess the applicability of deep learning in dna classification level. Since the training phase of DBN is costly expensive, specially if deals with DNA sequence with thousand of variables, we introduce a new encoding method, using decimal-binary vector to represent the sequence as input to the model, thereafter compare with one-hot-vector encoding in two datasets. We evaluated our proposed model with different contrastive algorithms which achieved significant improvement for the training speed with comparable classification result. This result has shown a potential of using decimal-binary vector on DBN for DNA sequence to solve other sequence problem in bioinformatics.

• Annual Conference on Human and Language Technology
• /
• 2022.10a
• /
• pp.360-365
• /
• 2022

문장 표현(sentence representation)은 자연어처리 분야 내의 다양한 문제 해결 및 응용 개발에 있어 유용하게 활용될 수 있는 주요한 도구 중 하나이다. 하지만 최근 널리 도입되고 있는 사전 학습 언어 모델(pre-trained language model)로부터 도출한 문장 표현은 이방성(anisotropy)이 뚜렷한 등 그 고유의 특성으로 인해 문장 유사도(Semantic Textual Similarity; STS) 측정과 같은 태스크에서 기대 이하의 성능을 보이는 것으로 알려져 있다. 이러한 문제를 해결하기 위해 대조 학습(contrastive learning)을 사전 학습 언어 모델에 적용하는 연구가 문헌에서 활발히 진행되어 왔으며, 그중에서도 레이블이 없는 데이터를 활용하는 비지도 대조 학습 방법이 주목을 받고 있다. 하지만 대다수의 기존 연구들은 주로 영어 문장 표현 개선에 집중하였으며, 이에 대응되는 한국어 문장 표현에 관한 연구는 상대적으로 부족한 실정이다. 이에 본 논문에서는 대표적인 비지도 대조 학습 방법(ConSERT, SimCSE)을 다양한 한국어 사전 학습 언어 모델(KoBERT, KR-BERT, KLUE-BERT)에 적용하여 문장 유사도 태스크(KorSTS, KLUE-STS)에 대해 평가하였다. 그 결과, 한국어의 경우에도 일반적으로 영어의 경우와 유사한 경향성을 보이는 것을 확인하였으며, 이에 더하여 다음과 같은 새로운 사실을 관측하였다. 첫째, 사용한 비지도 대조 학습 방법 모두에서 KLUE-BERT가 KoBERT, KR-BERT보다 더 안정적이고 나은 성능을 보였다. 둘째, ConSERT에서 소개하는 여러 데이터 증강 방법 중 token shuffling 방법이 전반적으로 높은 성능을 보였다. 셋째, 두 가지 비지도 대조 학습 방법 모두 검증 데이터로 활용한 KLUE-STS 학습 데이터에 대해 성능이 과적합되는 현상을 발견하였다. 결론적으로, 본 연구에서는 한국어 문장 표현 또한 영어의 경우와 마찬가지로 비지도 대조 학습의 적용을 통해 그 성능을 개선할 수 있음을 검증하였으며, 이와 같은 결과가 향후 한국어 문장 표현 연구 발전에 초석이 되기를 기대한다.

• Annual Conference on Human and Language Technology
• /
• 2022.10a
• /
• pp.262-267
• /
• 2022

이커머스 리뷰와 같은 특정 도메인의 경우, 텍스트 표현벡터 학습을 위한 양질의 오픈 학습 데이터를 구하기 어렵다. 또한 사람이 수동으로 검수하며 학습데이터를 만드는 경우, 많은 시간과 비용을 소모하게 된다. 따라서 본 논문에서는 수동으로 검수된 데이터없이 양질의 텍스트 표현벡터를 만들 수 있도록 두 단계의 대조 학습 시스템을 제안한다. 이 두 단계 대조 학습 시스템은 레이블링 된 학습데이터가 필요하지 않은 자기지도 학습 단계와 리뷰의 특성을 고려한 자동 레이블링 기반의 지도 학습 단계로 구성된다. 또한 노이즈에 강한 오류함수와 한국어에 유효한 데이터 증강 기법을 적용한다. 그 결과 스피어먼 상관 계수 기반의 성능 평가를 통해, 베이스 모델과 비교하여 성능을 14.03 향상하였다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.11a
• /
• pp.557-560
• /
• 2023

In this paper, we introduce a pre-training method leveraging the capabilities of the Vision Transformer (ViT) for disease diagnosis in conventional Fundus images. Recognizing the need for effective representation learning in medical images, our method combines the Vision Transformer with a Masked Autoencoder to generate meaningful and pertinent image augmentations. During pre-training, the Masked Autoencoder produces an altered version of the original image, which serves as a positive pair. The Vision Transformer then employs contrastive learning techniques with this image pair to refine its weight parameters. Our experiments demonstrate that this dual-model approach harnesses the strengths of both the ViT and the Masked Autoencoder, resulting in robust and clinically relevant feature embeddings. Preliminary results suggest significant improvements in diagnostic accuracy, underscoring the potential of our methodology in enhancing automated disease diagnosis in fundus imaging.

• Korean Educational Research Journal
• /
• v.40 no.1
• /
• pp.17-42
• /
• 2019

Modality is the psychological attitude of the speaker, which is comprised by the sentences used in every language. Modality can be broadly categorized as perceptional modality and obligatory modality. This study summarizes the previous related literatures and theoretical branches of Korean linguistic studies. The study also proposes and classifies a modal concept on the Korean language, which is aimed at aiding Chinese people who are studying Korean. It further describes characteristics and expressions of modality in both the Chinese and Korean languages. This study aims to develop an effective teaching-learning program on the basis of the contrastive analysis between Korean language's modality, "-(으)ㄹ 수 있다," and the corresponding Chinese auxiliary verb, "能." Modality is a syntax item that reflects a speaker's subjective manner. There are many grammatical facets in Korean language books and teaching materials that are modal in nature. Further, modalities in Korean language are not only numerous but also have very rich meanings and functions. Based on the contrastive analysis, this study designs an effective teaching plan for Chinese people learning the Korean language. The designed system uses specific conversational occasions as the basis of learning, and it adapts the Korean language's modal system to classroom teaching. The system is expected to be effective during classroom teaching for demonstrating and learning modality in the Korean language.

• The Journal of Korean Association of Computer Education
• /
• v.7 no.4
• /
• pp.93-102
• /
• 2004

For children with hearing impairment, unlike the children without hearing impairment who can reconstruct their languages through the process of hearing and uttering, the inherent mechanism for language acquisition do not operate due to the loss of hearing ability. Therefore, to help hearing-impaired children develop their language ability, web-based language learning system should be constructed depending on the special qualities which the children possess in language learning process. When the system is being designed, it is necessary that words or expressions describing actions or situations be animated and that active situation-based language learning system be constructed to help them develop their power of observation. Moreover, the system needs to be developed through the use of alternative thinking strategy, antonyms, and contrastive words, and emphasis on facial expressions. This paper presents web-based language learning system which is suitable for hearing-impaired children in the way to reduce the grammatical errors they make and to improve their language learning.

• Korean Journal of Radiology
• /
• v.24 no.11
• /
• pp.1061-1080
• /
• 2023

Artificial intelligence (AI) in radiology is a rapidly developing field with several prospective clinical studies demonstrating its benefits in clinical practice. In 2022, the Korean Society of Radiology held a forum to discuss the challenges and drawbacks in AI development and implementation. Various barriers hinder the successful application and widespread adoption of AI in radiology, such as limited annotated data, data privacy and security, data heterogeneity, imbalanced data, model interpretability, overfitting, and integration with clinical workflows. In this review, some of the various possible solutions to these challenges are presented and discussed; these include training with longitudinal and multimodal datasets, dense training with multitask learning and multimodal learning, self-supervised contrastive learning, various image modifications and syntheses using generative models, explainable AI, causal learning, federated learning with large data models, and digital twins.