• Annual Conference on Human and Language Technology
• /
• 2023.10a
• /
• pp.50-55
• /
• 2023

속성 기반 감정 분석은 텍스트 내에서 감정과 해당 감정이 특정 속성, 예를 들어 제품의 특성이나 서비스의 특징에 어떻게 연결되는지를 분석하는 태스크이다. 본 논문에서는 속성 기반 감정 분석 데이터를 사용한 다중 작업-토큰 레이블링 문제에 프롬프트 튜닝 기법을 적용하기 위한 포괄적인 방법론을 소개한다. 이러한 방법론에는 토큰 레이블링 문제를 시퀀스 레이블링 문제로 일반화하기 위한 감정 표현 영역 검출 파이프라인이 포함된다. 또한 분리된 시퀀스들을 속성과 감정에 대해 분류 하기 위한 템플릿을 선정하고, 데이터셋 특성에 맞는 레이블 워드를 확장하는 방법을 제안함으써 모델의 성능을 최적화한다. 최종적으로, 퓨샷 세팅에서의 속성 기반 감정 분석 태스크에 대한 몇 가지 실험 결과와 분석을 제공한다. 구축된 데이터와 베이스라인 모델은 AIHUB(www.aihub.or.kr)에 공개되어 있다.

• Korean Journal of Remote Sensing
• /
• v.38 no.2
• /
• pp.199-213
• /
• 2022

The unsupervised domain adaptation can solve the impractical issue of repeatedly collecting high-quality training data every year for annual crop classification. This study evaluates the applicability of deep learning-based unsupervised domain adaptation models for crop classification. Three unsupervised domain adaptation models including a deep adaptation network (DAN), a deep reconstruction-classification network, and a domain adversarial neural network (DANN) are quantitatively compared via a crop classification experiment using unmanned aerial vehicle images in Hapcheon-gun and Changnyeong-gun, the major garlic and onion cultivation areas in Korea. As source baseline and target baseline models, convolutional neural networks (CNNs) are additionally applied to evaluate the classification performance of the unsupervised domain adaptation models. The three unsupervised domain adaptation models outperformed the source baseline CNN, but the different classification performances were observed depending on the degree of inconsistency between data distributions in source and target images. The classification accuracy of DAN was higher than that of the other two models when the inconsistency between source and target images was low, whereas DANN has the best classification performance when the inconsistency between source and target images was high. Therefore, the extent to which data distributions of the source and target images match should be considered to select the best unsupervised domain adaptation model to generate reliable classification results.

• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.8
• /
• pp.325-330
• /
• 2022

This paper conducts the collection and experiment of datasets for deep learning research on sign language such as sign language recognition, sign language translation, and sign language segmentation for Korean sign language. There exist difficulties for deep learning research of sign language. First, it is difficult to recognize sign languages since they contain multiple modalities including hand movements, hand directions, and facial expressions. Second, it is the absence of training data to conduct deep learning research. Currently, KETI dataset is the only known dataset for Korean sign language for deep learning. Sign language datasets for deep learning research are classified into two categories: Isolated sign language and Continuous sign language. Although several foreign sign language datasets have been collected over time. they are also insufficient for deep learning research of sign language. Therefore, we attempted to collect a large-scale Korean sign language dataset and evaluate it using a baseline model named TSPNet which has the performance of SOTA in the field of sign language translation. The collected dataset consists of a total of 11,402 image and text. Our experimental result with the baseline model using the dataset shows BLEU-4 score 3.63, which would be used as a basic performance of a baseline model for Korean sign language dataset. We hope that our experience of collecting Korean sign language dataset helps facilitate further research directions on Korean sign language.

• The Journal of the Korea Contents Association
• /
• v.9 no.2
• /
• pp.27-35
• /
• 2009

In this paper, we present result of embedded system based H.264/SVC decoder circuit design and system implementation. To deal with the standardized H.264/SVC functionalities, the presented SVC decoder system is consist of hardware engine design and software with ARM core processor. In order to improve the feasibility and applicability, and reduce the decoder complexity, the implemented system is constructed with only the consideration of IPPP structure scalability without using the full B-picture architecture. Finally, we will show the decoding image result using the designed H.264/SVC decoder system.

• Annual Conference on Human and Language Technology
• /
• 2006.10e
• /
• pp.51-56
• /
• 2006

일반적으로 운율 정보를 음성인식에 이용한 연구들에 있어서는 대부분 운율의 음향적 정보를 이용하는데 반하여, 본 연구에서는 운율어나 음절수와 같은 운율의 구조적 정보가 인식률 향상에 기여함을 보인다. 본 논문은 두 가지 운율 정보, 즉 운율어와 음절수를 이용하여 발음모델링을 할 경우에 음성인식기의 성능을 평가하는 것을 목표로 하는 것으로, 먼저, 운율어를 이용하여 위치 정보데이터의 가능한 모든 발음을 생성하고, 다시 음절 수를 기준으로 발음변이 수를 조절하는 방법을 제시한 다음, 제안한 방법에 의하여 생성한 발음사전을 이용하여 음성인식의 성능을 평가하였다. 실험결과 운율어를 이용하여 발음 사전을 제작한 모든 경우에 베이스라인과 비교하여 성능이 향상됨을 보였는데, 베이스라인의 WER 4.63% 에서 최대 8.4%의 WER 가 감소하였다. 위치 정보 데이터의 음절수에 따라서 발음 변이의 수를 조절한 결과도 전체적으로는 3 음절로 그 수를 제한한 경우, 6 음절이상 단어에서는 4음절로 제한한 경우에 가장 좋은 인식 성능을 얻을 수 있어서, 음절수에 따른 발음변이 수의 조절이 효과적임을 알 수 있었다.

• Annual Conference on Human and Language Technology
• /
• 2020.10a
• /
• pp.347-350
• /
• 2020

농식품 가격을 안정적으로 제공하기 위해 농식품 가격 변동에 대한 요인 분석이 필요하다. 본 연구는 농식품 가격 변동의 요인 분석을 위해 인과관계 템플릿을 정의하고, 요약을 위한 개체명 인식 방법을 적용한다. 농식품 일일동향 데이터에 대한 평가에서 딥러닝 기반 BiLSTM-CRF 실험 결과 F1-점수 0.93으로 베이스라인 Bi-LSTM 실험 결과 0.75에 비해 높은 성능을 보였다.

• Journal of the Korea Society of Computer and Information
• /
• v.13 no.4
• /
• pp.1-9
• /
• 2008

Most consumer electronics companies hold numerous line-ups to cope with divergent customer's needs. To cope with current situation, most products are derived from the 'base product' which is developed for brand new features with respect to the change requests. That is called derivation. After 'base code' is developed for newly introduced products, some modification will occur corresponding to the derivative product models. So, quality attributes of 'base code' affects quality and productivity of 'derived code'. But in the middle of continuous modification to 'base code', violation of architectural design decision and unauthorized or maybe unsophisticated change to source code willing to happen and thus it cause critical problem. Those code has 'aging symptom' both architectural and code level in nature. In this paper, we introduced reengineering framework which guide the procedure and tactics to find and fix 'aging symptom' for improvement on quality attribute of 'base code'.

• Proceedings of the Korea Database Society Conference
• /
• 2002.10a
• /
• pp.494-510
• /
• 2002

Overview: web-based 기능제공. 가망고객/우수 고객들에 대한 수익 및 파이프 라인관리. Business Results: 영업력 향상. 향상된 영업관리 제고 (중략)

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.42 no.5
• /
• pp.437-443
• /
• 2014

Engine is one of the most important parts in a rocket for completing its mission successfully. In this paper, we provide a methodology for estimating reliability and development cost of a liquid rocket engine newly developed. To estimate reliability, a baseline engine is selected considering factors whose effects on reliability are unquantifiable. Then reliability of a baseline engine is adjusted to reflect the effect of factors that can be modeled quantitatively. Using the previous Transcost engine cost expressed in terms of mass and the number of hot firing tests, the engine development cost is reexpressed in reliability and thrust requirements. Finally, a numerical example is given to illustrate the application of the methodology to a turbopump rocket engine using staged combustion cycle with LOX/LH2 propellant.

• Journal of the Korean Society for information Management
• /
• v.18 no.3
• /
• pp.159-178
• /
• 2001

This paper presents an automatic text summarization model which selects representative sentences from sentence clusters to create a summary. Summary generation experiments were performed on two sets of test documents after learning the optimum environment from a training set. Centroid clustering method turned out to be the most effective in clustering sentences, and sentence weight was found more effective than the similarity value between sentence and cluster centroid vectors in selecting a representative sentence from each cluster. The result of experiments also proves that inverse sentence weight as well as title word weight for terms and location weight for sentences are effective in improving the performance of summarization.