• Journal of Korea Technology Innovation Society
• /
• v.22 no.3
• /
• pp.340-361
• /
• 2019

This study analyzes the technological structure of artificial intelligence (AI) and technological capabilities of AI companies based on patent information. 2589 AI patents registered in USPTO from 2007 to 2017 were collected and analyzed by the Latent Dirichlet Allocation (LDA) to derive 20 AI technology topics. Analysis of technology development trends by AI technology reveals that visual understanding, data analysis, motion control, and machine learning are growing, while language understanding and speech technology are sluggish. In addition, we also investigated leading companies in each sub-field of AI as well as core competencies of global IT companies. The findings of this study are expected to be fruitfully used for formulation and implementation of technology strategy of AI companies.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2021.06a
• /
• pp.448-448
• /
• 2021

최근 급격한 기후변화와 도시화 및 산업화로 인한 지류하천에서의 수량과 수질의 변동은 생물 다양성 감소와 수생태계 건강성 저하에 큰 영향을 미치고 있다. 효율적인 수생태 관리를 위해서는 지속적인 유량, 수질, 그리고 수생태 모니터링을 통한 데이터 축적과 더불어 면밀한 상관 분석을 통해 수생태계 건강성의 악화 원인을 규명해야 할 필요가 있다. 그러나 수많은 지류하천을 대상으로 한 지속적인 모니터링은 현실적으로 어려움이 있으며, 수생태계의 특성 상 단일 영향 인자만으로 수생태계의 건강성 변화와의 관계를 정확히 파악하는데 한계가 있다. 따라서 지류하천에서의 유량 및 수질의 시공간적인 변동성과 다양한 영향 인자를 고려하여 수생태계의 건강성을 효율적으로 예측할 수 있는 기술이 필요하다. 이에 본 연구에서는 경험적 데이터 기반의 머신러닝 모델 구축을 통해 미계측 하천에서의 수생태계 건강성 지수(BMI, TDI, FAI)의 등급(A to E)을 예측하고자 하였다. 머신러닝 모델은 학습 데이터셋의 양과 질에 따라 성능이 크게 달라질 수 있으며, 학습 데이터셋의 분포가 불균형적일 경우 과적합 또는 과소적합 문제가 발생할 수 있다. 이를 보완하고자 본 연구에서는 실제 측정망 데이터셋을 바탕으로 생성적 적대 신경망 GAN(Generative Adversarial Network) 알고리즘을 통해 머신러닝 모델 학습에 필요한 추가 데이터셋(유량, 수질, 기상, 수생태 등급)을 확보하였다. 머신러닝 모델의 성능은 5차 교차검증 과정을 통해 평가하였으며, GAN 데이터셋의 정확도는 실제 측정망 데이터셋의 정규분포와의 비교 분석을 통해 평가하였다. 최종적으로 SWAT(Soil and Water Assessment Tool) 모형을 통해 예측 된 미계측 하천에서의 데이터셋을 머신러닝 모델의 검증 자료로 사용하여 수생태계 건강성 등급 예측 정확도를 평가하였다. 본 연구에서의 GAN에 의해 강화된 머신러닝 모델은 수질 및 수생태 관리가 필요한 우심 지류하천 선정과 구조적/비구조적 최적관리기법에 따른 수생태계 건강성 개선 효과를 평가하는데 활용될 수 있을 것이다. 또한 이를 통해 예측된 미계측 하천에서의 수생태계 건강성 등급 자료는 수량-수질-수생태를 유기적으로 연계한 통합 물관리 정책을 수립하는데 기초자료로 활용될 수 있을 것이라 사료된다.

• The Korean Journal of Archival Studies
• /
• no.54
• /
• pp.211-248
• /
• 2017

This study examined ways to improve records management by using the new technology of the Fourth Industrial Revolution. To do this, we selected four technologies that have a huge impact on the production and management of records such as cloud, big data, artificial intelligence, and the Internet of Things. We tested these technologies and summarized their concepts, characteristics, and applications. The characteristics of the changed production records were analyzed by each technology. Because of new technology, the production of records has rapidly increased and the types of records have become diverse. With this, there is also a need for solutions to explain the quality of data and the context of production. To effectively introduce each technology into records management, a roadmap should be designed by classifying which technology should be applied immediately and which should be applied later depending on the maturity of the technology. To cope with changes in the characteristics of production records, a flexible data structure must be produced in a standardized format. Public authorities should also be able to procure Software as a Service (SaaS) products and use digital technology to improve the quality of public services.

• Journal of Korean Society for Quality Management
• /
• v.50 no.4
• /
• pp.647-664
• /
• 2022

Purpose: The purpose of this study is to derive major policies that domestic small and medium-sized manufacturing companies should consider to maximize productivity and quality improvement by utilizing manufacturing data and AI, and to find priorities and implications. Methods: In this study, domestic and international issues and literature review by country were conducted to derive major considerations such as manufacturing AI technology, manufacturing AI talent, manufacturing AI data and manufacturing AI ecosystem. Additionally, the questionnaire survey targeting 46 experts of manufacturing data and AI industry were conducted. Finally, the major considerations and detailed factors importance were derived by applying the Analytic Hierarchy Process (AHP). Results: As a result of the study, it was found that 'manufacturing AI technology', 'manufacturing AI talent', 'manufacturing AI data', and 'manufacturing AI ecosystem' exist as key considerations for domestic manufacturing AI. After empirical analysis, the importance of the four key considerations was found to be 'manufacturing AI ecosystem (0.272)', 'manufacturing AI data (0.265)', 'manufacturing AI technology (0.233)', and 'manufacturing AI talent (0.230)'. The importance of the derived four viewpoints is maintained at a similar level. In addition, looking at the detailed variables with the highest importance for each of the four perspectives, 'Best Practice', 'manufacturing data quality management regime, 'manufacturing data collection infrastructure', and 'manufacturing AI manpower level of solution providers' were found. Conclusion: For the sustainable growth of the domestic manufacturing AI ecosystem, it should be possible to develop and promote manufacturing AI policies in a balanced way by considering all four derived viewpoints. This paper is expected to be used as an effective guideline when developing policies for upgrading manufacturing through domestic manufacturing data and AI in the future.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.05a
• /
• pp.456-457
• /
• 2021

Recently, in order to more accurately provide indoor location-based services, technologies using Wi-Fi fingerprints and deep learning are being studied. Among the deep learning models, an RNN model that can store information from the past can store continuous movements in indoor positioning, thereby reducing positioning errors. When using an RNN model for indoor positioning, the collected training data must be continuous sequential data. However, the Wi-Fi fingerprint data collected to determine specific location information cannot be used as training data for an RNN model because only RSSI for a specific location is recorded. This paper proposes a region clustering technique for sequential input data generation of RNN models based on Wi-Fi fingerprint data.

• Journal of the Korean Society for Library and Information Science
• /
• v.53 no.2
• /
• pp.223-246
• /
• 2019

This research makes the metadata element design for Korean medicine research data management and re-use. Derived metadata elements are verified in research data of Korea Institute of Oriental Medicine. TTAK.K0-10.0976 Standard, DataCite metadata Schema and National Research Data Platform of KISTI were analyzed to derive the metadata elements. Including Identifier, 27 elements were derived as top-level elements with 29 mandatory elements, 13 recommended elements and 31 optional elements. The degree of elements' necessity and new metadata elements were investigated and suggested in the survey by six domain experts in korean medicine field. In this study subject classification for the korean medicine research data are suggested. The final version of metadata schema was tested and verified by comparing with the legacy metadata fields. The research results can be used to describe the Korean medicine research data: items and files.

• Journal of the Korean Society for information Management
• /
• v.40 no.4
• /
• pp.329-349
• /
• 2023

With the rapid development of information and communication technology, the speed of data production has increased rapidly, and this is represented by the concept of big data. Discussions on quality and reliability are also underway for big data whose data scale has rapidly increased in a short period of time. On the other hand, small data is minimal data of excellent quality and means data necessary for a specific problem situation. In the field of culture and arts, data of various types and topics exist, and research using big data technology is being conducted. However, research on whether basic information about culture and arts institutions is accurately provided and utilized is insufficient. The basic information of an institution can be an essential basis used in most big data analysis and becomes a starting point for identifying an institution. This study collected data dealing with the basic information of culture and arts institutions to define common metadata and constructed small data in the form of a knowledge graph linking institutions around common metadata. This can be a way to explore the types and characteristics of culture and arts institutions in an integrated way.

• Journal of the Korean Society for information Management
• /
• v.24 no.3
• /
• pp.179-200
• /
• 2007

This study identifies the interdisciplinary structures of 8 scientific disciplines in science and technology using the data from journal citations and web links, and compares the interdisciplinarity among these scientific disciplines. The interdisciplinarity refers to interdisciplinary connections among scientific fields and the degree of interdisciplinarity is measured by the number of associated fields and the rate of self-citation. A re-arranged classification scheme for science and technology was adopted to identify subject categories of journals and web pages. Web link analysis revealed a few additional interdisciplinary connections that were not identified by the journal citation analysis, thus demonstrating that it is useful means of investigating the interdisciplinarity of scientific fields. Besides, in most of the cases the interdisciplinarity of the engineering fields were found greater than that of the fields in natural sciences in both analyses.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2017.05a
• /
• pp.543-546
• /
• 2017

Recently, people are using the data analysis in order to follow the new trend in information and communication technology. Media plays an important role to expand the new issue in our society, especially affected to establish social awareness about science and technology. So, We find some major technologies (Machine Learning & Blockchains) of future communication and information based on the 200 news articles through two data analysis methods such as keyword analysis and sentiment analysis. We look forward this paper to constantly develop the technology of information and communication as the guiding frame of the new scientific world.

• Convergence Security Journal
• /
• v.24 no.2
• /
• pp.161-165
• /
• 2024

Recently, speech recognition models have been advancing, accompanied by the development of various speech processing technologies to obtain high-quality data. In the defense sector, efforts are being made to integrate technologies that effectively remove noise from speech data in noisy battlefield situations and enable efficient speech recognition. This paper proposes a method for effective speech recognition in the midst of diverse noise in a battlefield scenario, allowing commanders to convey orders. The proposed method involves noise removal from noisy speech followed by text conversion using OpenAI's Whisper model. Experimental results show that the proposed method reduces the Character Error Rate (CER) by 6.17% compared to the existing method that does not remove noise. Additionally, potential applications of the proposed method in the defense are discussed.