• Language and Information
• /
• v.6 no.2
• /
• pp.105-128
• /
• 2002

Morphologically annotated corpora are the basis for many tasks of computational linguistics. Most current approaches use statistically driven methods of morphological analysis, that provide just POS-tags. While this is sufficient for some applications, a rule-based full morphological analysis also yielding lemmatization and segmentation is needed for many others. This work thus aims at 〔1〕 introducing a rule-based Korean morphological analyzer called Kormoran based on the principle of linearity that prohibits any combination of left-to-right or right-to-left analysis or backtracking and then at 〔2〕 showing how it on be used as a POS-tagger by adopting an ordinary technique of preprocessing and also by filtering out irrelevant morpho-syntactic information in analyzed feature structures. It is shown that, besides providing a basis for subsequent syntactic or semantic processing, full morphological analyzers like Kormoran have the greater power of resolving ambiguities than simple POS-taggers. The focus of our present analysis is on Korean text.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2015.04a
• /
• pp.1001-1004
• /
• 2015

소설미디어의 급증과 함께 온라인 리뷰의 의존성이 급증하는 가운데 사용자의 올바른 의사결정을 저해하는 기만적 의견 스팸 이슈가 새롭게 주목받고 있다. 기존의 의견 스팸 연구는 실제 리뷰와 의견 스팸 간의 차이를 어휘, 품사 또는 감정단어와 같은 표면적 자질을 통해 설명하였으나 그들간의 의미적 연결관계는 고려하지 않았다. 본 논문에서는 1) 의미적 프레임 기반의 텍스트 분석기법을 제안하고, 이를 바탕으로 2) 의견 스팸과 실제 리뷰간의 의미적 차이가 있음을 규명하며 3) 새로운 의미적 프레임 자질을 사용하여 기존의 의견 스팸 분류 성능을 향상시킬 수 있음을 보인다.

• Genomics & Informatics
• /
• v.17 no.2
• /
• pp.13.1-13.3
• /
• 2019

The usage of controlled biomedical vocabularies is the cornerstone that enables seamless interoperability when using a common data model across multiple data sites. The Observational Health Data Science and Informatics (OHDSI) initiative combines over 100 controlled vocabularies into its own. However, the OHDSI vocabulary is limited in the sense that it combines multiple terminologies and does not provide a direct way to link them outside of their own self-contained scope. This issue makes the tasks of enriching feature sets by using external resources extremely difficult. In order to address these shortcomings, we have created a linked data version of the OHDSI vocabulary, connecting it with already established linked resources like bioportal, bio2rdf, etc. with the ultimate purpose of enabling the interoperability of resources previously foreign to the OHDSI universe.

• Journal of Information Processing Systems
• /
• v.15 no.5
• /
• pp.1119-1130
• /
• 2019

With the wide spread of Social Network Services (SNS), fake news-which is a way of disguising false information as legitimate media-has become a big social issue. This paper proposes a deep learning architecture for detecting fake news that is written in Korean. Previous works proposed appropriate fake news detection models for English, but Korean has two issues that cannot apply existing models: Korean can be expressed in shorter sentences than English even with the same meaning; therefore, it is difficult to operate a deep neural network because of the feature scarcity for deep learning. Difficulty in semantic analysis due to morpheme ambiguity. We worked to resolve these issues by implementing a system using various convolutional neural network-based deep learning architectures and "Fasttext" which is a word-embedding model learned by syllable unit. After training and testing its implementation, we could achieve meaningful accuracy for classification of the body and context discrepancies, but the accuracy was low for classification of the headline and body discrepancies.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2018.11a
• /
• pp.81-83
• /
• 2018

Fully Convolutional Network(FCN)은 기존의 방법보다 뛰어난 성능을 보였지만, FCN은 RGB 정보만을 사용하기 때문에 세밀한 예측이 필요한 장면에서는 다소 부족한 성능을 보였다. 이를 해결하기 위해 인코더-디코더 구조를 이용하여 RGB와 깊이의 멀티 모달을 활용하기 위한 FuseNet이 제안되었다. 하지만, FuseNet에서는 RGB와 깊이 브랜치 사이의 융합은 있지만, 인코더와 디코더 사이의 특징 지도를 융합하지 않는다. 본 논문에서는 FCN의 디코더 부분의 업샘플링 과정에서 이전 계층의 결과와 2배 업샘플링한 결과를 융합하는 스킵 레이어를 적용하여 FuseNet의 모달리티를 잘 활용하여 성능을 개선했다. 본 실험에서는 NYUDv2와 SUNRGBD 데이터 셋을 사용했으며, 전체 정확도는 각각 77%, 65%이고, 평균 IoU는 47.4%, 26.9%, 평균 정확도는 67.7%, 41%의 성능을 보였다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.6
• /
• pp.2115-2127
• /
• 2021

Although significant progress has been made in synthesizing visually realistic face images by Generative Adversarial Networks (GANs), there still lacks effective approaches to provide fine-grained control over the generation process for semantic facial attribute editing. In this work, we propose a novel cross channel self-attention based generative adversarial network (CCA-GAN), which weights the importance of multiple channels of features and archives pixel-level feature alignment and conversion, to reduce the impact on irrelevant attributes while editing the target attributes. Evaluation results show that CCA-GAN outperforms state-of-the-art models on the CelebA dataset, reducing Fréchet Inception Distance (FID) and Kernel Inception Distance (KID) by 15~28% and 25~100%, respectively. Furthermore, visualization of generated samples confirms the effect of disentanglement of the proposed model.

• Journal of information and communication convergence engineering
• /
• v.19 no.1
• /
• pp.36-41
• /
• 2021

Text-based image retrieval is not only cumbersome as it requires the manual input of keywords by the user, but is also limited in the semantic approach of keywords. However, content-based image retrieval enables visual processing by a computer to solve the problems of text retrieval more fundamentally. Vision applications such as extraction and mapping of image characteristics, require the processing of a large amount of data in a mobile environment, rendering efficient power consumption difficult. Hence, an effective image retrieval method on mobile platforms is proposed herein. To provide the visual meaning of keywords to be inserted into images, the efficiency of image retrieval is improved by extracting keywords of exchangeable image file format metadata from images retrieved through a content-based similar image retrieval method and then adding automatic keywords to images captured on mobile devices. Additionally, users can manually add or modify keywords to the image metadata.

• Journal of Information Processing Systems
• /
• v.17 no.6
• /
• pp.1138-1156
• /
• 2021

Emotional semantics are the highest level of semantics that can be extracted from an image. Constructing a system that can automatically recognize the emotional semantics from images will be significant for marketing, smart healthcare, and deep human-computer interaction. To understand the direction of image emotion recognition as well as the general research methods, we summarize the current development trends and shed light on potential future research. The primary contributions of this paper are as follows. We investigate the color, texture, shape and contour features used for emotional semantics extraction. We establish two models that map images into emotional space and introduce in detail the various processes in the image emotional semantic recognition framework. We also discuss important datasets and useful applications in the field such as garment image and image retrieval. We conclude with a brief discussion about future research trends.

• The Journal of Korea Robotics Society
• /
• v.17 no.3
• /
• pp.334-338
• /
• 2022

In visual simultaneous localization and mapping (vSLAM), the correct recognition of a place benefits in relocalization and improved map accuracy. However, its performance is significantly affected by the environmental conditions such as variation in light, viewpoints, seasons, and presence of dynamic objects. This research addresses the problem of feature occlusion caused by interference of dynamic objects leading to the poor performance of visual place recognition algorithm. To overcome the aforementioned problem, this research analyzes the role of scene semantics in correct detection of a place in challenging environments and presents a semantics aided visual place recognition method. Semantics being invariant to viewpoint changes and dynamic environment can improve the overall performance of the place matching method. The proposed method is evaluated on the two benchmark datasets with dynamic environment and seasonal changes. Experimental results show the improved performance of the visual place recognition method for vSLAM.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2024.05a
• /
• pp.794-797
• /
• 2024

Despite advancements in deep learning, existing semantic segmentation models exhibit suboptimal performance under adverse weather conditions, such as fog or rain, whereas they perform well in clear weather conditions. To address this issue, much of the research has focused on making image or feature-level representations weather-independent. However, disentangling the style and content of images remains a challenge. In this work, we propose a novel fine-tuning method, 'freeze-n-update.' We identify a subset of model parameters that are weather-independent and demonstrate that by freezing these parameters and fine-tuning others, segmentation performance can be significantly improved. Experiments on a test dataset confirm both the effectiveness and practicality of our approach.