• Design Convergence Study
• /
• v.16 no.5
• /
• pp.125-140
• /
• 2017

This study explores potential capability of a machine learning approach for monitoring urban morphology based on an evident case study. The case study conveys year 2006 investigations on interpreting urban morphology of Boston Main Streets by applying a machine learning approach. From the lesson of the precedent study, in 2016, another field research and interview was conducted to compare changes in urban situation, data commons culture, and technology innovation during the decade. This paper describes open possibilities to advance urban monitoring for morphological changes. Most of all, a multi-participatory data platform enables managing urban data system in real time. Second, collaboration with machines with artificial intelligence can intervene the framework of the urban management system as well as transform it through new demands of innovative industries. Recently, urban regeneration became a dominant urban planning strategy in Korean, therefore, urban monitoring is on demand. It is timely important to correspond to in-situ problems based on empirical research.

• Tunnel and Underground Space
• /
• v.33 no.4
• /
• pp.281-298
• /
• 2023

Continuous research efforts are being devoted to unmanned mobile platforms for lunar exploration. There is an ongoing demand for real-time information processing to accurately determine the positioning and mapping of areas of interest on the lunar surface. To apply deep learning processing and analysis techniques to practical rovers, research on software integration and optimization is imperative. In this study, a foundational investigation has been conducted on real-time analysis of virtual lunar base construction site images, aimed at automatically quantifying spatial information of key objects. This study involved transitioning from an existing region-based object recognition algorithm to a boundary box-based algorithm, thus enhancing object recognition accuracy and inference speed. To facilitate extensive data-based object matching training, the Batch Hard Triplet Mining technique was introduced, and research was conducted to optimize both training and inference processes. Furthermore, an improved software system for object recognition and identical object matching was integrated, accompanied by the development of visualization software for the automatic matching of identical objects within input images. Leveraging satellite simulative captured video data for training objects and moving object-captured video data for inference, training and inference for identical object matching were successfully executed. The outcomes of this research suggest the feasibility of implementing 3D spatial information based on continuous-capture video data of mobile platforms and utilizing it for positioning objects within regions of interest. As a result, these findings are expected to contribute to the integration of an automated on-site system for video-based construction monitoring and control of significant target objects within future lunar base construction sites.

• Journal of the Korea Society for Simulation
• /
• v.17 no.2
• /
• pp.31-47
• /
• 2008

As OWL(Web Ontology Language) has been selected as a standard ontology description language by W3C, many ontologies have been building and developing in OWL. The lena developed by HP as an Application Programming Interface(API) provides various APIs to develop inference engines as well as storages, and it is widely used for system development. However, the storage model of Jena2 stores most owl documents not acceptable into a single table and it shows low processing performance for a large ontology data set. Most of all, Jena2 storage model does not consider hierarchical structures of classes and properties. In addition, it shows low query processing performance using the hierarchical structure because of many join operations. To solve these issues, this paper proposes an OWL ontology relational database model. The proposed model semantically classifies and stores information such as classes, properties, and instances. It improves the query processing performance by managing hierarchical information in a separate table. This paper also describes the implementation and evaluation results. This paper also shows the experiment and evaluation result and the comparative analysis on both results. The experiment and evaluation show our proposal provides a prominent performance as against Jena2.

• Journal of the Korea Society of Computer and Information
• /
• v.10 no.1 s.33
• /
• pp.93-100
• /
• 2005

This study is to desist and implement an efficient microbial medical image retrieval system based on knowledge and content of them which can make use of more accurate decision on colony as doll as efficient education for new techicians. For this. re first address overall inference to set up flexible search path using rule-base in order U redure time required original microbial identification by searching the fastest path of microbial identification phase based on heuristics knowledge. Next, we propose a color ffature gfraction mtU, which is able to extract color feature vectors of visual contents from a inn microbial image based on especially bacteria image using HSV color model. In addition, for better retrieval performance based on large microbial databases, we present an integrated indexing technique that combines with B+-tree for indexing simple attributes, inverted file structure for text medical keywords list, and scan-based filtering method for high dimensional color feature vectors. Finally. the implemented system shows the possibility to manage and retrieve the complex microbial images using knowledge and visual contents itself effectively. We expect to decrease rapidly Loaming time for elementary technicians by tell organizing knowledge of clinical fields through proposed system.

• Journal of the Korea Society for Simulation
• /
• v.18 no.2
• /
• pp.29-38
• /
• 2009

In the computing environment with heterogeneous resources, a job scheduling model is necessary for effective resource utilization and high-speed data processing. And, the job scheduling model has to cope with a dynamic change in the condition of resources. There have been lots of researches on resource estimation methods and heuristic algorithms about how to distribute and allocate jobs to heterogeneous resources. But, existing researches have a weakness for system compatibility and scalability because they do not support the standard language. Also, they are impossible to process jobs effectively and deal with a variety of computing situations in which the condition of resources is dynamically changed in real-time. In order to solve the problems of existing researches, this paper proposes a semantic computing-based dynamic job scheduling model that defines various knowledge-based rules for job scheduling methods adaptable to changes in resource condition and allocate a job to the best suited resource through inference. This paper also constructs a resource ontology to manage information about heterogeneous resources without difficulty as using the OWL, the standard ontology language established by W3C. Experimental results shows that the proposed scheduling model outperforms existing scheduling models, in terms of throughput, job loss, and turn around time.

• Phonetics and Speech Sciences
• /
• v.15 no.3
• /
• pp.75-82
• /
• 2023

Automatic speech recognition (ASR) has been revolutionized with deep learning-based approaches, among which self-supervised learning methods have proven to be particularly effective. In this study, we aim to enhance the performance of OpenAI's Whisper model, a multilingual ASR system on the Korean language. Whisper was pretrained on a large corpus (around 680,000 hours) of web speech data and has demonstrated strong recognition performance for major languages. However, it faces challenges in recognizing languages such as Korean, which is not major language while training. We address this issue by fine-tuning the Whisper model with an additional dataset comprising about 1,000 hours of Korean speech. We also compare its performance against a Transformer model that was trained from scratch using the same dataset. Our results indicate that fine-tuning the Whisper model significantly improved its Korean speech recognition capabilities in terms of character error rate (CER). Specifically, the performance improved with increasing model size. However, the Whisper model's performance on English deteriorated post fine-tuning, emphasizing the need for further research to develop robust multilingual models. Our study demonstrates the potential of utilizing a fine-tuned Whisper model for Korean ASR applications. Future work will focus on multilingual recognition and optimization for real-time inference.

• Journal of Intelligence and Information Systems
• /
• v.28 no.2
• /
• pp.127-146
• /
• 2022

Recently, as word embedding has shown excellent performance in various tasks of deep learning-based natural language processing, researches on the advancement and application of word, sentence, and document embedding are being actively conducted. Among them, cross-language transfer, which enables semantic exchange between different languages, is growing simultaneously with the development of embedding models. Academia's interests in vector alignment are growing with the expectation that it can be applied to various embedding-based analysis. In particular, vector alignment is expected to be applied to mapping between specialized domains and generalized domains. In other words, it is expected that it will be possible to map the vocabulary of specialized fields such as R&D, medicine, and law into the space of the pre-trained language model learned with huge volume of general-purpose documents, or provide a clue for mapping vocabulary between mutually different specialized fields. However, since linear-based vector alignment which has been mainly studied in academia basically assumes statistical linearity, it tends to simplify the vector space. This essentially assumes that different types of vector spaces are geometrically similar, which yields a limitation that it causes inevitable distortion in the alignment process. To overcome this limitation, we propose a deep learning-based vector alignment methodology that effectively learns the nonlinearity of data. The proposed methodology consists of sequential learning of a skip-connected autoencoder and a regression model to align the specialized word embedding expressed in each space to the general embedding space. Finally, through the inference of the two trained models, the specialized vocabulary can be aligned in the general space. To verify the performance of the proposed methodology, an experiment was performed on a total of 77,578 documents in the field of 'health care' among national R&D tasks performed from 2011 to 2020. As a result, it was confirmed that the proposed methodology showed superior performance in terms of cosine similarity compared to the existing linear vector alignment.