• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.05a
• /
• pp.493-495
• /
• 2023

Deep learning methods and attention mechanisms have been incorporated to improve facial emotion recognition, which has recently attracted much attention. The fusion approaches have improved accuracy by combining various types of information. This research proposes a fusion network with self-attention and local attention mechanisms. It uses a multi-layer perceptron network. The network extracts distinguishing characteristics from facial images using pre-trained models on RAF-DB dataset. We outperform the other fusion methods on RAD-DB dataset with impressive results.

• International Journal of Advanced Culture Technology
• /
• v.12 no.1
• /
• pp.128-133
• /
• 2024

Painting, as an artistic creation with aesthetic characteristics, is inseparable from "shape". "Modeling" is a subjective way of expressing objects, and "black shape" is not only a way of expression, but also a way of composition. "Black shape" is a way of expressing the overall picture of the creator, and "black shape" extracted from the pattern analysis of black and white gray structure is a powerful means to deeply explore the emotions and thoughts of the painter. How to create the expression of subjective intention in the creation of artificial figure painting and to explore the beauty of picture form and the presentation of formal language in multiple dimensions is always sought by painters. Tang Dynasty political stability, economic and cultural development of unprecedented prosperity, "Toujian Diu" as a representative painting of Tang Dynasty, the figure is plump and realistic, colorful but elegant. It not only represents the aesthetic taste of the Tang Dynasty, but also reflects the overall style of the Tang Dynasty's artificial figure painting. The aim of this paper is to explore the value of "black shape" in "Toughen Diagram" and analyze its multi-component form in depth.

• Journal of Information Processing Systems
• /
• v.20 no.1
• /
• pp.1-12
• /
• 2024

Nowadays, Islamic content is widely used in research, including Hadith and the Al-Quran. Both are mostly used in the field of natural language processing, especially in text classification research. One of the difficulties in learning the Al-Quran is ambiguity, while the Al-Quran is used as the main source of Islamic law and the life guidance of a Muslim in the world. This research was proposed to relieve people in learning the Al-Quran. We proposed a word embedding feature-based on Tensor Space Model as feature extraction, which is used to reduce the ambiguity. Based on the experiment results and the analysis, we prove that the proposed method yields the best performance with the Hamming loss 0.10317.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2024.01a
• /
• pp.335-337
• /
• 2024

This paper presents a new open-world object detection method emphasizing uncertainty representation in machine learning models. The focus is on adapting to real-world uncertainties, incrementally updating the model's knowledge repository for dynamic scenarios. Applications like autonomous vehicles benefit from improved multi-class classification accuracy. The paper reviews challenges in existing methodologies, stressing the need for universal detectors capable of handling unknown classes. Future directions propose collaboration, integration of language models, to improve the adaptability and applicability of open-world object detection.

• Journal of Korean Academy of Nursing
• /
• v.41 no.4
• /
• pp.558-567
• /
• 2011

Purpose: The purpose of this study was to monitor the use of health-related quality of life (HRQOL) instruments in Korean studies of patients with diabetes. Methods: Of 86 Korean studies initially identified, 17 studies met the inclusion criteria. For each study, a description of the instrument and its psychometric properties were monitored by the Instrument Review Criteria of the Scientific Advisory Committee. These criteria include conceptual definition, attributes, taxonomy, reliability, validity, responsiveness, administrative mode, and language adaptations. Results: Five generic and one diabetes specific type questionnaires were identified from the 17 studies. Of those studies, conceptual definitions with the attributes of multi-dimension and subjectiveness were provided for 11 studies (71%). In the analysis of conceptual taxonomy, only 6 studies were classified as HRQOL, while other studies were done as QOL or health status. In monitoring of psychometric properties, reliability, validity, and responsiveness were reported for 88.2%, 64.7%, and 29.4%, respectively. One generic instrument was developed with a Korean population, while the other instruments were developed for Western countries. However, language adaptations were performed for only a few of the instruments. Conclusion: The psychometric properties including responsiveness of most instruments warrants further research, and the development of diabetes-specific HRQOL measurements should be sought to facilitate intervention outcomes across Korean studies of patients with diabetes.

• The KIPS Transactions:PartB
• /
• v.11B no.1
• /
• pp.35-44
• /
• 2004

Web contents created for desktop screens give rise to problems when they are to be displayed on the small screens of mobile terminals. While in some cases some of the objects of a page may not be displayable due to the lack of browser capability, the entire page may not be displayable due to the incompatibility with the browser in other cases. In this paper, we introduce a new mobile content adaptation approach based on web clipping, which transforms an original page into one that is optimally displayed on a mobile terminal. In this method, a source page is automatically clipped and transformed according to the clip specification made by the content provider using a clip editing tool. The clip editing tool allows the user to specify group clips, multi-level cups and dynamic clips as well as simple clips, and the presentation layout through a graphic user interface. Based on the clip specifications, each clip is transformed into an intermediate meta-language document, which in turn is transformed into a presentation page in the target markup language. Transcoding of image objects in major image file formats is also supported.

• The Transactions of the Korea Information Processing Society
• /
• v.4 no.1
• /
• pp.255-269
• /
• 1997

In this paper, we describe the implementation of C cross-compiler for the ES-C2340 DSP2 processor by using the GNU compiler. For the rapid and efficient developing of the compiler and other parts like the processor-dependent back -end which is implemented newly to build the compiler. This approach has several advantages. First, as we use GNU compiler's well-proved excellent optimization method and multi-language support capability, we can improve he efficiency and generality of the compiler. Second, as we concentrate on the high-level language as logic approving tool in processor developing process. And to support the cross-compiler, we also implement a text-level pre-linker.

• Journal of KIISE
• /
• v.44 no.4
• /
• pp.383-391
• /
• 2017

As we enter a new era of Big Data, the amount of semantic data has rapidly increased. In order to derive meaningful information from this large semantic data, studies that utilize the SWRL(Semantic Web Rule Language) are being actively conducted. SWRL rules are based on data extracted from a user's empirical knowledge. However, conventional reasoning systems developed on single machines cannot process large scale data. Similarly, multi-node based reasoning systems have performance degradation problems due to network shuffling. Therefore, this paper overcomes the limitations of existing systems and proposes more efficient distributed inference methods. It also introduces data partitioning strategies to minimize network shuffling. In addition, it describes a method for optimizing the incremental reasoning process through data selection and determining the rule order. In order to evaluate the proposed methods, the experiments were conducted using WiseKB consisting of 200 million triples with 83 user defined rules and the overall reasoning task was completed in 32.7 minutes. Also, the experiment results using LUBM bench datasets showed that our approach could perform reasoning twice as fast as MapReduce based reasoning systems.

• Annual Conference on Human and Language Technology
• /
• 2018.10a
• /
• pp.400-405
• /
• 2018

본 연구는 한국어 화장품 리뷰 코퍼스의 자질기반 감성 분석을 위하여, 이 도메인에서 실현되는 중요한 다단어 표현(MWE)의 유한상태 그래프 사전과 문법을 구축하는 방법론을 제시하고, 실제 구축된 사전과 문법의 성능을 평가하는 것을 목표로 한다. 본 연구에서는 자연어처리(NLP)에서 중요한 화두로 논의되어 온 MWE의 어휘-통사적 특징을 부분문법 그래프(LGG)로 형식화하였다. 화장품 리뷰 코퍼스에 DECO 한국어 전자사전을 적용하여 어휘 빈도 통계를 획득하고 이에 대한 언어학적 분석을 통해 극성 MWE(Polarity-MWE)와 화제 MWE(Topic MWE)의 전체 네 가지 하위 범주를 분류하였다. 또한 각 모듈간의 상호관계에 대한 어휘-통사적 속성을 반복적으로 적용하는 이중 증식(double-propagation)을 통해 자원을 확장하였다. 이 과정을 통해 구축된 대용량 MWE 유한그래프 사전 DECO-MWE의 성능을 테스트한 결과 각각 0.844(Pol-MWE), 0.742(Top-MWE)의 조화평균을 보였다. 이를 통해 본 연구에서 제안하는 MWE 언어자원 구축 방법론이 다양한 도메인에서 활용될 수 있고 향후 자질기반 감성 분석에 중요한 자원이 될 것임을 확인하였다.

• Journal of The Korean Astronomical Society
• /
• v.47 no.3
• /
• pp.115-122
• /
• 2014

The New Vacuum Solar Telescope (NVST) is a 1-meter vacuum solar telescope that aims to observe the fine structures of active regions on the Sun. The main tasks of the NVST are high resolution imaging and spectral observations, including the measurements of the solar magnetic field. The NVST has been collecting more than 20 million FITS files since it began routine observations in 2012 and produces maximum observational records of 120 thousand files in a day. Given the large amount of files, the effective archiving and retrieval of files becomes a critical and urgent problem. In this study, we implement a new data archiving system for the NVST based on the Fastbit Not Only Structured Query Language (NoSQL) database. Comparing to the relational database (i.e., MySQL; My Structured Query Language), the Fastbit database manifests distinctive advantages on indexing and querying performance. In a large scale database of 40 million records, the multi-field combined query response time of Fastbit database is about 15 times faster and fully meets the requirements of the NVST. Our slestudy brings a new idea for massive astronomical data archiving and would contribute to the design of data management systems for other astronomical telescopes.