• Phonetics and Speech Sciences
• /
• v.7 no.3
• /
• pp.89-97
• /
• 2015

The aim of this study is to investigate the prosodic aspects of structural ambiguous sentences in Korean produced by Japanese Korean learners and the influence of their first language prosody. Previous studies reported that structural ambiguous sentences in Korean are different especially in prosodic phrasing. So we examined whether Japanese Korean leaners can also distinguish, in production, between two types of structural ambiguous sentences on the basis of prosodic features. For this purpose 4 Korean native speakers and 8 Japanese Korean learners participated in the production test. Analysis materials are 6 sentences where a relative clause modify either NP1 or NP1+NP2. The results show that Korean native speakers produced ambiguous sentences by different prosodic structure depending on their semantic and syntactic structure (left branching or right branching sentence). Japanese speakers also show distinct prosodic structure for two types of ambiguous sentences in most cases, but they have more errors in producing left branching sentences than right branching sentences. In addition to that, interference of Japanese pitch accent in the production of Korean ambiguous sentences was observed.

• Journal of the Korea Society of Computer and Information
• /
• v.19 no.11
• /
• pp.105-113
• /
• 2014

In this paper, we propose an OWL DL mapping rules for construction legal ontology based on the analyzed relationship between the structural features and elements of the statute. The mapping rule to be proposed is the method building the structure of the domestic statute, unique attribute of the statute, and reference relation between laws with TBox, and the legal sentence is analyzed, and the pattern type of the sentence is selected. It expresses with ABox. The proposed mapping rule is transformed to the information in which the computer can process the domestic legal document. It is usable for the legal knowledge base.

• Journal of the Korean Society of Food Culture
• /
• v.36 no.2
• /
• pp.168-183
• /
• 2021

Using the big data analysis of the Choseonwangjosilrok, this research aimed to figure out the fruits' types, prevalence, seasonal appearances as well as the royalty's perspective on fruits during Choseon period. Choseonwangjosilrok included nineteen kinds of fruits and five kinds of nuts, totaling 1,601 cases at 72.8% and 533 cases at 24.2% respectively. The text recorded fruits being used as: tributes for kings, gifts from kings to palace officials, tomb offerings, county specialties, trade goods or gifts to the foreign ambassadors, and medicine ingredients in oriental pharmacy. Seasonally the fruits appeared demonstrating an even distribution. Periodic characteristics were observed in decreasing quantity chronologically. From fifteenth century to nineteenth century, the fruits with timely features were seen: 804 times at 36.6%, 578 times at 26.3%, 490 times at 22.3%, 248 times at 11.3%, and 78 times at 3.5% respectively. In fifteenth century: citrons, quinces, pomegranates, cherries, permissions, watermelons, Korean melons, omija, walnuts, chestnuts, and pine nuts appeared most frequently. In sixteenth century: pears, grapes, apricots, peaches, and hazelnuts appeared most frequently. In seventeenth century: tangerines and dates appeared most frequently. In eighteenth century, trifoliate orange was the most frequently mentioned fruit.

• Korean Journal of Remote Sensing
• /
• v.37 no.1
• /
• pp.111-122
• /
• 2021

Building footprint extraction is an active topic in the domain of remote sensing, since buildings are a fundamental unit of urban areas. Deep convolutional neural networks successfully perform footprint extraction from optical satellite images. However, semantic segmentation produces coarse results in the output, such as blurred and rounded boundaries, which are caused by the use of convolutional layers with large receptive fields and pooling layers. The objective of this study is to generate visually enhanced building objects by directly extracting the vertices of individual buildings by combining instance segmentation and keypoint detection. The target keypoints in building extraction are defined as points of interest based on the local image gradient direction, that is, the vertices of a building polygon. The proposed framework follows a two-stage, top-down approach that is divided into object detection and keypoint estimation. Keypoints between instances are distinguished by merging the rough segmentation masks and the local features of regions of interest. A building polygon is created by grouping the predicted keypoints through a simple geometric method. Our model achieved an F1-score of 0.650 with an mIoU of 62.6 for building footprint extraction using the OpenCitesAI dataset. The results demonstrated that the proposed framework using keypoint estimation exhibited better segmentation performance when compared with Mask R-CNN in terms of both qualitative and quantitative results.

• Journal of the Architectural Institute of Korea Planning & Design
• /
• v.35 no.2
• /
• pp.93-104
• /
• 2019

Minimal art, which began to flourish in the mid-1960s, explores perceptual situations caused by the involvement of objects in given site contexts. This has led to the mentions of minimal art as a site-specific art, but its limitations have also been pointed out. This study specifically addresses the limitations of minimal art as a site-specific art with two perceptual points of view. First, according to Michael Fried, situations described as 'now here' focus largely on the bodily experiences of a place. However, they do not rooted in specific time and space of a certain place. Second, the unique characteristics of a certain place are excluded from the perception of the body which occupies the passage of time. Self-sufficient algorithm, which is far from site-specific conditions, is the autonomous system creating the period in the way of arrangement of objects. In addition, Minimal art regards a body only as the objectivity excluding the subjectivity which is essential creating meaning in a place. In the latter part of the article, these features are dealt with through Donald Judd's works. This study on site-specificity also provides a new perspective on the discussion of Minimal architecture and Minimal landscape.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.4
• /
• pp.1275-1292
• /
• 2021

As a fine-grained classification problem, aspect-level sentiment classification predicts the sentiment polarity for different aspects in context. To address this issue, researchers have widely used attention mechanisms to abstract the relationship between context and aspects. Still, it is difficult to effectively obtain a more profound semantic representation, and the strong correlation between local context features and the aspect-based sentiment is rarely considered. In this paper, a hybrid attention capsule network for aspect-level sentiment classification (ABASCap) was proposed. In this model, the multi-head self-attention was improved, and a context mask mechanism based on adjustable context window was proposed, so as to effectively obtain the internal association between aspects and context. Moreover, the dynamic routing algorithm and activation function in capsule network were optimized to meet the task requirements. Finally, sufficient experiments were conducted on three benchmark datasets in different domains. Compared with other baseline models, ABASCap achieved better classification results, and outperformed the state-of-the-art methods in this task after incorporating pre-training BERT.

• Journal of KIBIM
• /
• v.9 no.4
• /
• pp.31-40
• /
• 2019

With Building Information Modeling(BIM) becoming the de facto standard for data sharing in the AEC industry, additional needs have increased to ensure the data integrity of BIM models themselves. Although the Industry Foundation Classes provide an open and neutral data format, its generalized schema leaves it open to data loss and misclassifications This research applied deep learning to automatically classify BIM elements and thus check the integrity of BIM-to-IFC mappings. Multi-view CNN(MVCC) and PointNet, which are two deep learning models customized to learn and classify in 3 dimensional non-euclidean spaces, were used. The analysis was restricted to classifying subtypes of architectural walls. MVCNN resulted in the highest performance, with ACC and F1 score of 0.95 and 0.94. MVCNN unitizes images from multiple perspectives of an element, and was thus able to learn the nuanced differences of wall subtypes. PointNet, on the other hand, lost many of the detailed features as it uses a sample of the point clouds and perceived only the 'skeleton' of the given walls.

• Journal of Korean Biological Nursing Science
• /
• v.23 no.3
• /
• pp.170-179
• /
• 2021

Purpose: The aim of this study was to identify core keywords and topic groups in the 'Diabetes mellitus and mobile applications' field of research for better understanding research trends in the past 20 years. Methods: This study was a text-mining and topic modeling study including four steps such as 'collecting abstracts', 'extracting and cleaning semantic morphemes', 'building a co-occurrence matrix', and 'analyzing network features and clustering topic groups'. Results: A total of 789 papers published between 2002 and 2021 were found in databases (Springer). Among them, 435 words were extracted from 118 articles selected according to the conditions: 'analyzed by text network analysis and topic modeling'. The core keywords were 'self-management', 'intervention', 'health', 'support', 'technique' and 'system'. Through the topic modeling analysis, four themes were derived: 'intervention', 'blood glucose level control', 'self-management' and 'mobile health'. The main topic of this study was 'self-management'. Conclusion: While more recent work has investigated mobile applications, the highest feature was related to self-management in the diabetes care and prevention. Nursing interventions utilizing mobile application are expected to not only effective and powerful glycemic control and self-management tools, but can be also used for patient-driven lifestyle modification.

• ETRI Journal
• /
• v.44 no.3
• /
• pp.413-425
• /
• 2022

The study of literary texts is one of the earliest disciplines practiced around the globe. Poetry is artistic writing in which words are carefully chosen and arranged for their meaning, sound, and rhythm. Poetry usually has a broad and profound sense that makes it difficult to be interpreted even by humans. The essence of poetry is Rasa, which signifies mood or emotion. In this paper, we propose a poetry classification-based approach to automatically extract similar poems from a repository. Specifically, we perform a novel Rasa-based classification of Hindi poetry. For the task, we primarily used lexical features in a bag-of-words model trained using the support vector machine classifier. In the model, we employed Hindi WordNet, Latent Semantic Indexing, and Word2Vec-based neural word embedding. To extract the rich feature vectors, we prepared a repository containing 37 717 poems collected from various sources. We evaluated the performance of the system on a manually constructed dataset containing 945 Hindi poems. Experimental results demonstrated that the proposed model attained satisfactory performance.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.3
• /
• pp.771-791
• /
• 2022

Protein-protein interaction (PPI) extraction from original text is important for revealing the molecular mechanism of biological processes. With the rapid growth of biomedical literature, manually extracting PPI has become more time-consuming and laborious. Therefore, the automatic PPI extraction from the raw literature through natural language processing technology has attracted the attention of the majority of researchers. We propose a PPI extraction model based on the large pre-trained language model and adversarial training. It enhances the learning of semantic and syntactic features using BioBERT pre-trained weights, which are built on large-scale domain corpora, and adversarial perturbations are applied to the embedding layer to improve the robustness of the model. Experimental results showed that the proposed model achieved the highest F1 scores (83.93% and 90.31%) on two corpora with large sample sizes, namely, AIMed and BioInfer, respectively, compared with the previous method. It also achieved comparable performance on three corpora with small sample sizes, namely, HPRD50, IEPA, and LLL.