• International Journal of Advanced Culture Technology
• /
• 제12권2호
• /
• pp.167-179
• /
• 2024

This study addresses the challenge of objectively evaluating the performance of early-stage startups amidst limited information and uncertainty. Focusing on companies selected by TIPS, a leading private sector-driven startup support policy in Korea, the research develops a new indicator to assess technological efficiency. By analyzing various input and output variables collected from Crunchbase and KIND (Korea Investor's Network for Disclosure System) databases, including technology use metrics, patents, and Crunchbase rankings, the study derives technological efficiency for TIPS-selected startups. A prediction model is then developed utilizing machine learning techniques such as Random Forest and boosting (XGBoost) to classify startups into efficiency percentiles (10th, 30th, and 50th). The results indicate that prediction accuracy improves with higher percentiles based on the technical efficiency index, providing valuable insights for evaluating and predicting startup performance in early markets characterized by information scarcity and uncertainty. Future research directions should focus on assessing growth potential and sustainability using the developed classification and prediction models, aiding investors in making data-driven investment decisions and contributing to the development of the early startup ecosystem.

• International Journal of Advanced Culture Technology
• /
• 제12권3호
• /
• pp.148-158
• /
• 2024

We investigate Autism Spectrum Disorder (ASD), which is typified by deficits in social interaction, repetitive behaviors, limited vocabulary, and cognitive delays. Traditional diagnostic methodologies, reliant on expert evaluations, frequently result in deferred detection and intervention, particularly in South Korea, where there is a dearth of qualified professionals and limited public awareness. In this study, we employ advanced deep learning algorithms to enhance early ASD screening through automated video analysis. Utilizing architectures such as Convolutional Long Short-Term Memory (ConvLSTM), Long-term Recurrent Convolutional Network (LRCN), and Convolutional Neural Networks with Gated Recurrent Units (CNN+GRU), we analyze video data from platforms like YouTube and TikTok to identify stereotypic behaviors (arm flapping, head banging, spinning). Our results indicate that the LRCN model exhibited superior performance with 79.61% accuracy on the augmented platform video dataset and 79.37% on the original SSBD dataset. The ConvLSTM and CNN+GRU models also achieved higher accuracy than the original SSBD dataset. Through this research, we underscore AI's potential in early ASD detection by automating the identification of stereotypic behaviors, thereby enabling timely intervention. We also emphasize the significance of utilizing expanded datasets from social media platform videos in augmenting model accuracy and robustness, thus paving the way for more accessible diagnostic methods.

• 대한가정학회지
• /
• 제29권3호
• /
• pp.11-21
• /
• 1991

The study has been intended to find out meaningful information about the development of a prototype of enhanced maternity girdle. The girdles of three different models which were available in the market have been carried out by three six-month pregnant women and three nine-month pregnant women. The results of the study are as follows. 1. All three girdles showed improved wear effects in order of model A, model B, model C. Body surface area measurement and two body surface angles of abdomen are significantly dicreased by wearing and type of girdles. It is presumed that the reasons of good wear effect of model A is low expansion rate of the material and tight fitness of the model. Model B is made of material whose expansion rate is higher than model A. Also abdominal part of the model B is bias cut which is considered to result better stretch and consequently lower wear effect. 2. For wear comfort, subjects preferred in order of model B, model C, and model A. All subjects feel more comfortable after wearing girdles 30 minutes than after wearing girdles 1 day. Comparing 2 subject groups, 6-month pregnant group feel more comfortable about wearing girdles than 9-month pregnant group. 3. The girdles are expanded as a whole in order of model B, model C and model A. Considering the expansion rate of some specific area of the girdles, abdominal area expands more than hip area which expands more than thigh area. The expansion rates of girdles worn to 6-month pregnant group are very low at all area, while the expansion rates of girdles worn to 9-month pregnant group are very high.

• Journal of Korea Trade
• /
• 제25권1호
• /
• pp.65-88
• /
• 2021

Purpose - This paper aims to examine the determinants of SWFs' investment in Asian countries and to identify consistent investment patterns of SWFs in specific target firms from Asia, particularly China and South Korea. Design/methodology - This study extends the Tobin's Q model to examine the relationship between SWF investments in target firms and their returns with other firm-level control variables. We collect consistent data on SWF investments and the matched firm-level data on target firms, which of observation is 1,512 firms (333 in South Korea and 1,179 in China) targeted by 20 SWF sources during 1997-2017. The panel random effect model is used to estimate the extended Tobin's Q model. The robustness of the estimations is tested by the simultaneous equation models and the panel GEE model. Findings - The evidence shows that sovereign wealth funds are more inclined to invest in the financial sector with a monopoly position and in large firms with higher growth opportunity and superior cash asset ratios in China. In contrast to their investments in China, sovereign wealth funds in South Korea prefer to invest in strategic sectors, such as energy and information technology, and in large firms with high performance and low leverage. Sovereign wealth funds' investments tend to significantly improve the target firm's performance measured by sales growth and returns in both Korea and China. Originality/value - The existing literature focuses on examining the determination of SWFs investment in the developed countries, such as Europe and the United States. Our paper contributes to the literature in three ways; first, we analyzes case studies of SWF investments in Asian markets, which are less developed and riskier. Second, we examine whether the determination of SWF investment in Asian target firms depends on the different time periods, on types of sources of SWFs, and on acquiring countries. Third, our research uses vast sample data on target firms in longer time periods (1997-2017) than other previous studies on the SWFs for Asian markets.

• 산업경영시스템학회지
• /
• 제38권2호
• /
• pp.129-137
• /
• 2015

The study aims at quantifying the effect of nano technology in the fields of economics and social aspects by using the methodology of system dynamics. A case study which using selenium oxide nanoparticles as additive agent in order to enhance fuel efficiency was selected as an example of nano technology in economic and societal benefits. Additionally, models for exhaust gas from combustion of fuel (diesel) and related issues are developed to evaluate real-time assessment of the effect of nano technology. It was found that the selenium oxide nanoparticles increase fuel efficiency, and it also affects on the amount of exhaust gas and the respiratory disease related issues. The results of this study which give quantitative value for the effect of nano technology can be used as objective references in development of national policy.

• 지적과 국토정보
• /
• 제50권1호
• /
• pp.181-200
• /
• 2020

본 연구는 OLS모형을 적용하여 주택보유기간에 영향을 미치는 결정요인을 추정한 후 SVM, Decision Tree, Random Forest, Gradient Boosting, XGBoost, LightGBM을 통해 각 모형별 예측력을 비교하였다. 예측력이 가장 높은 모델을 기반모델 삼아 앙상블 모형 중 하나인 Stacking모형을 적용하여 더욱 예측력이 높은 모형을 구축하여 주택시장의 주택거래량을 파악할 수 있다는 점에 선행 연구와의 차이가 있다. OLS분석 결과 매도이익, 주택가격, 가구원 수, 거주주택형태(단독주택, 아파트)이 주택보유기간에 영향을 미치는 것으로 나타났으며, RMSE를 기준삼아 각 머신러닝 모형과 예측력 비교한 결과 머신러닝 모델의 예측력이 더 높은 것으로 나타났다. 이후, 영향을 미치는 변수로 데이터를 재구축한 후 각 머신러닝을 적용하여 예측력을 비교하였으며, 분석 결과 Random Forest의 예측력이 가장 우수한 것으로 나타났다. 또한 예측력이 가장 높은 Random Forest, Decision Tree, Gradient Boosting, XGBoost모형을 개별모형으로 적용하고, Linear, Ridge, Lasso모형을 메타모델로 하여 Stacking 모형을 구축하였다. 분석 결과, Ridge모형일 때 RMSE값이 0.5181으로 가장 낮게 나타나 예측력이 가장 높은 모델을 구축하였다.

• 한국정보시스템학회지:정보시스템연구
• /
• 제13권2호
• /
• pp.22-48
• /
• 2004

Information systems that are not used cannot be useful. In order to increase user acceptance, it is necessary to understand why people accept or reject information systems. Technology Acceptance Model(TAM) is one of the most influential research models for studying determinants how users accept information systems. Recently, Knowledge Management Systems(KMS) have become important components of corporate systems as the foundation of industrialized economics has shifted from natural resources to knowledge assets. This paper applies TAM to investigate users' acceptance of KMS in public administration institutions. It sampled 182 users who had experience in using KMS. Many empirical researches have suggested that TAM can be integrated with other organizational theories to improve its predictive and explanatory ower. We extended the basic TAM by the integration of appraisal and reward satisfaction theory. There are many external variables that influence the perception and the belief of system users. We introduced two external variables(job characteristics, IT self-efficacy) and one additional perception variable, perceived appraisal and reward(PAR) in the basic TAM model. The LISREL model analysis is used for finding out the causality among variables and testing the model fitness. As result, The IT self-efficacy influences to the perceived ease of use(PEOU) and the PAR, and the PEOU influences directly to the perceived usefulness(PU), the PAR, and the attitude toward KMS. The KMS participation intention(PI) was influenced by the PAR and the attitude directly,andbythePEOUindirectly. Finally, this paper suggests some guidelines for the adoption of KMS in public sectors on the basis of the study results.

• 한국농림기상학회지
• /
• 제3권3호
• /
• pp.156-162
• /
• 2001

In order to monitor local climatic information, twelve automated weather stations (AWS) were installed in alpine area by the Alpine Agricultural Experiment Station, Rural Development Administration (RDA), at the field of major crop located in around highland area, and collected data from 1993 to 2000. Hourly measurements of air and soil temperature (underground 10 cm,20 cm), relative humidity, wind speed and direction, precipitation, solar radiation and leaf wetness were automatically performed and the data could be collected through a public phone line. Datalogger was selected as CR10X (Campbell scientific, LTD, USA) out of consideration for sensers' compatibility, economics, endurance and conveniences. All AWS in alpine area were combined for net work and daily climatic data were analyzed in text and graphic file by program (Chumsungdae, LTD) on 1 km $\times$ 1 km grid tell basis. In this analysis system, important multi-functionalities, monitoring and analysis of local climatic information in alpine area was emphasized. The first objective was to obtain the output of a real time data from AWS. Secondly, daily climatic normals for each grid tell were calculated from geo-statistical relationships based on the climatic records of existing weather stations as well as their topographical informations. On 1 km $\times$ 1 km grid cell basis, real time climatic data from the automated weather stations and daily climatic normals were analyzed and graphed. In the future, if several simulation models were developed and connected with this system it would be possible to precisely forecast crop growth and yield or plant disease and pest by using climatic information in alpine area.

• 자원ㆍ환경경제연구
• /
• 제21권4호
• /
• pp.921-945
• /
• 2012

조건부 로짓(Conditional Logit: CL)모형은 모형추정 및 결과해석이 비교적 용이하다는 장점으로 널리 사용되는 반면에 응답자의 선호이질성(preference heterogeneity)을 충분히 반영하지 못한다는 한계를 가지고 있다. 본 연구에서는 최근 지배적인 방식으로 제시되고 있는 두개의 모형인 혼합로짓(Mixed Logit; ML)모형과 잠재계층모형(Latent Class Model; LCM)을 사용하여 우리나라 보령댐 주변 습지보호에 있어서 응답자간 선호이질성을 설명하고자 하였다. 6대광역시와 보령시 표본에 대하여 응답자별 이질성의 존재여부를 검토한 결과 두 지역간 뚜렷한 차이가 있음을 알 수 있었다. 즉 보령시의 경우에는 응답자간 선호이질성이 뚜렷하게 나타난데 반하여 6대광역시의 경우 응답자간 선호이질성이 거의 나타나지 않았다. 따라서 6대광역시의 경우에는 모수추정을 위해 CL 모형의 사용이 가능하나 보령시의 경우에는 선호이질성을 반영하기 위해 ML모형이나 LCM에 근거한 모수추정이 요구된다. 선호이질성의 원인을 규명하기 위하여 교차항이 있는 혼합로짓모형과 잠재계층모형을 고려할 수 있다. 교차항이 있는 혼합로짓모형의 경우 관찰되지 않은 개인단위의 이질성을 설명할 수 있는 장점이 있다. 그러나 두 모형을 비교한 결과 LCM이 교차항이 있는 ML모형이 제공하지 않는 추가적인 정보를 보여주는 것으로 나타나고 있다. 따라서 본 연구에서의 응답자간 선호이질성은 혼합로짓모형에 의한 개인적인 수준보다 잠재계층모형에 의한 계층단위에서 더 잘 설명될 수 있다고 할 수 있다.

• 한국산림과학회지
• /
• 제104권2호
• /
• pp.248-253
• /
• 2015

본 연구는 단기소득임산물에 대한 최적 생산계획을 수립하기 위한 의사결정 정보를 제공하는 수단으로 포트폴리오 기법을 적용하였다. 대상품목은 밤, 대추, 호두, 떫은감이며, 2008년부터 2013년까지 임산물생산비 통계의 생산량, 생산비, 조수입 자료를 이용하였으며 단위당 순수익을 분석하였다. 포트폴리오 모델에서 목적함수는 투자위험을 나타내는 유실수 품목의 수익 변동폭을 최소화하는 것이고, 제약조건은 최소 기대수익률을 달성하는 것이다. 분석결과, 2013년 유실수의 생산비율 밤 7%, 대추 20%, 호두 5%, 떫은감 68%과 비교하여 포트폴리오에서는 미래의 유실수 생산비율을 평균 밤 10%, 대추 9%, 호두 3%, 떫은감 78%로 구성하는 것이 안정적인 소득증대를 위해서 보다 효율적인 것으로 제시되었다. 이러한 원인은 호두와 대추가 순수익이 가장 많았지만 생산량과 수익의 등락폭이 상대적으로 컸고, 떫은감과 밤은 상대적으로 안정된 생산량과 일정한 수익을 유지하고 있기 때문인 것으로 분석되었다.