• Korean Journal of Remote Sensing
• /
• v.22 no.2
• /
• pp.123-129
• /
• 2006

High-resolution aerial color image offers great possibilities for geometric and semantic information for spatial data generation. However, shadow casts by buildings and trees in high-density urban areas obscure much of the information in the image giving rise to potentially inaccurate classification and inexact feature extraction. Though many researches have been implemented for solving shadow casts, few studies have been carried out about the extraction of features hindered by shadows from aerial color images in urban areas. This paper presents a asphalt road boundary extraction technique that combines information from aerial color image and LIDAR (LIght Detection And Ranging) data. The following steps have been performed to remove shadow effects and to extract road boundary from the image. First, the shadow regions of the aerial color image are precisely located using LEAR DSM (Digital Surface Model) and solar positions. Second, shadow regions assumed as road are corrected by shadow path reconstruction algorithms. After that, asphalt road boundary extraction is implemented by segmentation and edge detection. Finally, asphalt road boundary lines are extracted as vector data by vectorization technique. The experimental results showed that this approach was effective and great potential advantages.

• Journal of Cadastre & Land InformatiX
• /
• v.52 no.1
• /
• pp.37-55
• /
• 2022

This study presents the plans to derive quality items for develop the data quality standard for ensuring the quality of 3D building geospatial data in NDT(National Digital Twin). This paper is organized as follows. The first section briefly examines various factors that impact the quality of 3D geospatial data, and proposes the role and necessity of the data quality standard as a means of addressing the data errors properly and also meeting the minimum requirements of stakeholders. The second section analyzes the relationship between the standards - building data model for NDT and ISO 19157: Geospatial data quality - in order to consider directly relevant standards. Finally, we suggest three plans on developing NDT data quality standard: (1) the scope for evaluating data quality, (2) additional quality elements(geometric integrity, geometric fidelity, positional accuracy and semantic classification accuracy), and (3) NDT data quality items model based on ISO 19157. The plans reveled through the study would contribute to establish a way for the national standard on NDT data quality as well as the other standards associated with NDT over the coming years.

• International conference on construction engineering and project management
• /
• 2020.12a
• /
• pp.157-166
• /
• 2020

Automated rebar fabrication, which requires effective information exchange between model designers and fabricators, has brought the integration and interoperability of data from different sources to the notice of both academics and industry practitioners. Industry Foundation Classes (IFC) was one of the most commonly used data formats to represent the semantic information of prefabricated components in buildings, whereas the data format utilized by rebar fabrication machine is BundesVereinigung der Bausoftware (BVBS), which is a numerical data structure exchanging reinforcement information through ASCII encoded files. Seamless transformation between IFC and BVBS empowers the automated rebar fabrication and improve the construction productivity. In order to improve data interoperability between IFC and BVBS, this study presents an IFC extension based on the attributes required by automated rebar fabrication machines with the help of Information Delivery Manual (IDM) and Model View Definition (MVD). IDM is applied to describe and display the information needed for the design, construction and operation of projects, whereas MVD is a subset of IFC schema used to describe the automated rebar fabrication workflow. Firstly, with a rich pool of vocabularies practitioners, OmniClass is used in information exchange between IFC and BVBS, providing a hierarchy classification structure for reinforcing elements. Then, using International Framework for Dictionaries (IFD), the usage of each attribute is defined in a more consistent manner to assist the data mapping process. Besides, in order to address missing information within automated fabrication process, a schematic data mapping diagram has been made to deliver IFC information from BIM models to BVBS format for better data interoperability among different software agents. A case study based on the data mapping will be presented to demonstrate the proposed IFC extension and how it could assist/facilitate the information management.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.23 no.2
• /
• pp.131-156
• /
• 1998

Cooperative query answering is a research effort to develop a fault-tolerant and intelligent database system using the semantic knowledge base constructed from the underlying database. Such knowledge base has two aspects of usage. One is supporting the cooperative query answering Process for providing both an exact answer and neighborhood information relevant to a query. The other is supporting ongoing maintenance of the knowledge base for accommodating the changes in the knowledge content and database usage purpose. Existing studies have mostly focused on the cooperative query answering process but paid little attention on the dynamic knowledge base maintenance. This paper proposes a multi-level knowledge representation framework called Knowledge Abstraction Hierarchy (KAH) that can not only support cooperative query answering but also permit dynamic knowledge maintenance. The KAH consists of two types of knowledge abstraction hierarchies. The value abstraction hierarchy is constructed by abstract values that are hierarchically derived from specific data values in the underlying database on the basis of generalization and specialization relationships. The domain abstraction hierarchy is built on the various domains of the data values and incorporates the classification relationship between super-domains and sub-domains. On the basis of the KAH, a knowledge abstraction database is constructed on the relational data model and accommodates diverse knowledge maintenance needs and flexibly facilitates cooperative query answering. In terms of the knowledge maintenance, database operations are discussed for the cases where either the internal contents for a given KAH change or the structures of the KAH itself change. In terms of cooperative query answering, database operations are discussed for both the generalization and specialization Processes, and the conceptual query handling. A prototype system has been implemented at KAIST that demonstrates the usefulness of KAH in ordinary database application systems.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.4
• /
• pp.1101-1121
• /
• 2024

Thorough understanding of user demands and formulation of product development strategies are crucial in product design, and can effectively stimulate consumer behavior. Scientific categorization and classification of demands contribute to accurate design development, design efficiency, and success rates. In recent years, e-commerce has become important consumption platforms for smart wearable products. However, there are few studies on product design and development among those related to promoting platform product services and sales. Meanwhile, design strategies focusing on real consumer needs are scarce among smart wearable product design studies. Therefore, an empirical consumer demand analysis method is proposed and design development strategies are formulated based on a categorized interpretation of demands. Using representative smart bracelets from wearable smart products as a case, this paper classifies consumer demands with three methods: big data semantic analysis, KANO model analysis, and satisfaction analysis. The results reveal that analysis methods proposed herein can effectively classify consumer demands and confirm that differences in consumer demand categories have varying impacts on consumer behavior. On this basis, corresponding design strategies are proposed based on four categories of consumer demands, aiming to make product design the leading factor and promote consumer behavior on e-commerce platforms. This research further enriches demand research on smart wearable products on e-commerce platforms, and optimizes products from a design perspective, thereby promoting consumption. In future research, different data analysis methods will be tried to compare and analyze changes in consumer demands and influencing factors, thus improving research on impact factors of product design in e-commerce.

• Korean Journal of Cognitive Science
• /
• v.10 no.2
• /
• pp.27-34
• /
• 1999

The main purpose of this paper is to propose a general dialogue flow in 'the a appointment scheduling dialogues' in German using the concept of dialogue acts. A basic a assumption of this research is that dialogue acts contribute to the improvement of a translation system. They might be very useful to solve the problems that syntactic and semantic module could not resolve using contextual knowledge. The classification of the dialogue acts was conducted as a work of VERBMOBIL project and was based on real dialogues transcribed by experts. The real dialogues were analyzed in terms of the dialogue acts. We empirically analyzed the sequences of the dialogue acts not only in a series of dialogue turns but also in one dialogue turn. We attempted to analyZe the sequences in one dialogue turn additionally because the dialogue data used in this research showed some difference from the ones in other existing researches. By examining the sequences in dialogue acts. we proposed the dialogue flowchart in 'the a appointment scheduling dialogues' 'Based on the statistical analysis of the sequences of the most frequent dialogue acts. the dialogue flowcharts seem to represent' the a appointment scheduling dialogues' in general. A further research is required on c classification of dialogue acts which was a base for the analysis of dialogues. In order to e extract the most generalized model. we did not subcategorize each dialogue acts and used a limited number of items of dialogue acts. However. generally defined dialogue acts need to be defined more concretely and new dialogue acts for specific situations should be a added.

• Journal of Intelligence and Information Systems
• /
• v.26 no.1
• /
• pp.1-21
• /
• 2020

With the rapid development of artificial intelligence technology, various techniques have been developed to extract meaningful information from unstructured text data which constitutes a large portion of big data. Over the past decades, text mining technologies have been utilized in various industries for practical applications. In the field of business intelligence, it has been employed to discover new market and/or technology opportunities and support rational decision making of business participants. The market information such as market size, market growth rate, and market share is essential for setting companies' business strategies. There has been a continuous demand in various fields for specific product level-market information. However, the information has been generally provided at industry level or broad categories based on classification standards, making it difficult to obtain specific and proper information. In this regard, we propose a new methodology that can estimate the market sizes of product groups at more detailed levels than that of previously offered. We applied Word2Vec algorithm, a neural network based semantic word embedding model, to enable automatic market size estimation from individual companies' product information in a bottom-up manner. The overall process is as follows: First, the data related to product information is collected, refined, and restructured into suitable form for applying Word2Vec model. Next, the preprocessed data is embedded into vector space by Word2Vec and then the product groups are derived by extracting similar products names based on cosine similarity calculation. Finally, the sales data on the extracted products is summated to estimate the market size of the product groups. As an experimental data, text data of product names from Statistics Korea's microdata (345,103 cases) were mapped in multidimensional vector space by Word2Vec training. We performed parameters optimization for training and then applied vector dimension of 300 and window size of 15 as optimized parameters for further experiments. We employed index words of Korean Standard Industry Classification (KSIC) as a product name dataset to more efficiently cluster product groups. The product names which are similar to KSIC indexes were extracted based on cosine similarity. The market size of extracted products as one product category was calculated from individual companies' sales data. The market sizes of 11,654 specific product lines were automatically estimated by the proposed model. For the performance verification, the results were compared with actual market size of some items. The Pearson's correlation coefficient was 0.513. Our approach has several advantages differing from the previous studies. First, text mining and machine learning techniques were applied for the first time on market size estimation, overcoming the limitations of traditional sampling based- or multiple assumption required-methods. In addition, the level of market category can be easily and efficiently adjusted according to the purpose of information use by changing cosine similarity threshold. Furthermore, it has a high potential of practical applications since it can resolve unmet needs for detailed market size information in public and private sectors. Specifically, it can be utilized in technology evaluation and technology commercialization support program conducted by governmental institutions, as well as business strategies consulting and market analysis report publishing by private firms. The limitation of our study is that the presented model needs to be improved in terms of accuracy and reliability. The semantic-based word embedding module can be advanced by giving a proper order in the preprocessed dataset or by combining another algorithm such as Jaccard similarity with Word2Vec. Also, the methods of product group clustering can be changed to other types of unsupervised machine learning algorithm. Our group is currently working on subsequent studies and we expect that it can further improve the performance of the conceptually proposed basic model in this study.

• KIPS Transactions on Software and Data Engineering
• /
• v.12 no.10
• /
• pp.455-460
• /
• 2023

Text augmentation is a methodology that creates new augmented texts by transforming or generating original texts for the purpose of improving the performance of NLP models. However existing text augmentation techniques have limitations such as lack of expressive diversity semantic distortion and limited number of augmented texts. Recently text augmentation using large language models and few-shot learning can overcome these limitations but there is also a risk of noise generation due to incorrect generation. In this paper, we propose a text augmentation method called TAGS that generates multiple candidate texts and selects the appropriate text as the augmented text. TAGS generates various expressions using few-shot learning while effectively selecting suitable data even with a small amount of original text by using contrastive learning and similarity comparison. We applied this method to task-oriented chatbot data and achieved more than sixty times quantitative improvement. We also analyzed the generated texts to confirm that they produced semantically and expressively diverse texts compared to the original texts. Moreover, we trained and evaluated a classification model using the augmented texts and showed that it improved the performance by more than 0.1915, confirming that it helps to improve the actual model performance.

• Journal of Korean Tunnelling and Underground Space Association
• /
• v.24 no.6
• /
• pp.583-598
• /
• 2022

As human-based tunnel inspections are affected by the subjective judgment of the inspector, making continuous history management difficult. There is a lot of deep learning-based automatic crack detection research recently. However, the large public crack datasets used in most studies differ significantly from those in tunnels. Also, additional work is required to build sophisticated crack labels in current tunnel evaluation. Therefore, we present a method to improve crack detection performance by inputting existing datasets into a deep learning model. We evaluate and compare the performance of deep learning models trained by combining existing tunnel datasets, high-quality tunnel datasets, and public crack datasets. As a result, DeepLabv3+ with Cross-Entropy loss function performed best when trained on both public datasets, patchwise classification, and oversampled tunnel datasets. In the future, we expect to contribute to establishing a plan to efficiently utilize the tunnel image acquisition system's data for deep learning model learning.

• Korean Journal of Remote Sensing
• /
• v.39 no.6_3
• /
• pp.1679-1692
• /
• 2023

South Korea is a country that emits a large amount of pollutants as a result of population growth and industrial development and is also severely affected by transboundary air pollution due to its geographical location. As pollutants from both domestic and foreign sources contribute to air pollution in Korea, the location of air pollutant emission sources is crucial for understanding the movement and distribution of pollutants in the atmosphere and establishing national-level air pollution management and response strategies. Based on this background, this study aims to effectively acquire spatial information on domestic and international air pollutant emission sources, which is essential for analyzing air pollution status, by utilizing high-resolution optical satellite images and deep learning-based image segmentation models. In particular, industrial parks and quarries, which have been evaluated as contributing significantly to transboundary air pollution, were selected as the main research subjects, and images of these areas from multi-purpose satellites 3 and 3A were collected, preprocessed, and converted into input and label data for model training. As a result of training the U-Net model using this data, the overall accuracy of 0.8484 and mean Intersection over Union (mIoU) of 0.6490 were achieved, and the predicted maps showed significant results in extracting object boundaries more accurately than the label data created by course annotations.