• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.10
• /
• pp.2768-2787
• /
• 2023

As an important research direction of the application of computer science in the medical field, the automatic generation technology of radiology report has attracted wide attention in the academic community. Because the proportion of normal regions in radiology images is much larger than that of abnormal regions, words describing diseases are often masked by other words, resulting in significant feature loss during the calculation process, which affects the quality of generated reports. In addition, the huge difference between visual features and semantic features causes traditional multi-modal fusion method to fail to generate long narrative structures consisting of multiple sentences, which are required for medical reports. To address these challenges, we propose a multi-feature optimization Transformer (MFOT) for generating radiology reports. In detail, a multi-dimensional mapping attention (MDMA) module is designed to encode the visual grid features from different dimensions to reduce the loss of primary features in the encoding process; a feature pre-fusion (FP) module is constructed to enhance the interaction ability between multi-modal features, so as to generate a reasonably structured radiology report; a detail enhanced attention (DEA) module is proposed to enhance the extraction and utilization of key features and reduce the loss of key features. In conclusion, we evaluate the performance of our proposed model against prevailing mainstream models by utilizing widely-recognized radiology report datasets, namely IU X-Ray and MIMIC-CXR. The experimental outcomes demonstrate that our model achieves SOTA performance on both datasets, compared with the base model, the average improvement of six key indicators is 19.9% and 18.0% respectively. These findings substantiate the efficacy of our model in the domain of automated radiology report generation.

• Journal of Korea Port Economic Association
• /
• v.25 no.3
• /
• pp.139-164
• /
• 2009

The purpose of this paper is to investigate the value chain efficiency of Korean port investment by using the newly developed multi-year and multi-stage value chain efficiency model of DEA(Data Envelopment Analysis). Inputs[port investment amount, cargo handling capacity, and berthing capacity], and outputs[cargo handling amount, number of ship calls, revenue, and score of customer service satisfaction] are used during 14 years(1994-2007) for 20 Korean seaports by using two kinds of DEA models. Empirical main results are as follows: First, Model 1 shows that the ranking order of multi-stage value chain efficiency is Stage 2, Stage 3-1, Stage 1, and Stage 3-2. And according to the value chain average efficiency scores, ranking order is stages 2, 1, 3-1, and 3-2. In Model 2, 3(Incheon, Mogpo, and Jeju) out of 9 ports show the ranking order of Stages 2, 3-2, 3-1, and 1. And value chain average efficiency scores rank in order of Stages 2, 3-2, 3-1, and 1. Second, the difference among the value chain efficiency scores of each stage comes from the efficiency deterioration of all ports except Stages 2 and 1 in Model 1. In Model 2, value chain efficiency scores among the Stages 3-1, 3-2 compared to Stage 1 were deteriorated. The main policy implication based on the findings of this study is that the manager of port investment and management of Ministry of Land, Transport and Maritime Affairs in Korea should introduce the multi-year, multi-stage value chain efficiency method for deciding the port investment amount and evaluating the effect of port investment after considering the empirical results of this paper carefully.

• The Journal of the Korea Contents Association
• /
• v.18 no.10
• /
• pp.443-452
• /
• 2018

With the explosively increasing intellectual property rights, securing technological competitiveness of companies is more and more important. In particular, since patents include core technologies and element technologies, patent analysis researches are actively conducted to measure the technological value of companies. Various patent analysis studies have been conducted by the International Patent Classification(IPC), which does not include the latest technical classification, and the technical classification accuracy is low. In order to overcome this problem, the Cooperative Patent Classification(CPC), which includes the latest technology classification and detailed technical classification, has been developed. In this paper, we propose a model to analyze the classification of the technologies included in the patent by using the detailed classification system of CPC. It is possible to analyze the inventor's patents in consideration of the relation, importance, and efficiency between the detailed classification schemes of the CPCs to extract the core technology fields and to analyze the details more accurately than the existing IPC-based methods. Also, we perform the comparative evaluation with the existing IPC based patent analysis method and confirm that the proposed model shows better performance in analyzing the inventor's core technology classification.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2008.05a
• /
• pp.400-404
• /
• 2008

댐 군 연계운영을 위한 기존의 많은 최적화 모형은 경제성에 기반을 둔 단일 목적 함수를 가정함으로써 현실과는 동떨어진 결과를 도출하곤 하였다. 따라서 보다 현실적인 최적화 모형이 되기 위해서는 실제 운영과정을 모사할 수 있도록 적절한 초기 가중치를 부여하여 모형을 구축하고, 상충되는 목적간의 절충안으로 파레토 프런티어(Pareto-frontier)를 제시할 수 있는 다중목적 의사결정 기법이 요구된다. 본 연구의 목적은 댐 군 연계 운영을 위한 최적화 모형으로 소개된 CoMOM(Coordinated Multi-reservoir Operating Model)의 다중목적함수에 적합한 초기 가중치를 도출할 수 있는 시스템을 제안하는 것이다. 본 연구에서는 최적화 모형에 적합한 가중치를 결정함에 있어 댐의 초기저수량과 미래의 예상 유입량과 같은 수문 조건을 감안할 필요가 있음에 주목하였다. 이것은 초기저수량과 미래에 예상되는 유입량이 작을 경우 가급적 저수에 중점을 두고, 그 반대일 경우는 발전방류에 주안점을 두는 것이 바람직하다는 사실에서 추정해 볼 수 있는 가정이다. 따라서 댐의 초기 저수량 조건과 유입량 시나리오의 다양한 수문 조건별로 가장 적합한 가중치를 찾아본 후, 수문 조건을 입력요소로, 최적 가중치를 출력요소로 갖는 신경망 모형을 활용해서 수문 조건에 적합한 가중치를 예측할 수 있는 절차를 제안한다. 이 과정에서 최적 가중치를 도출하는 것이 관건이 될 수 있는데, 이를 위해 전승목 (2008)등이 제안한 DEA기반 순위결정 절차를 활용해서 최선의 파레토 최적해와 이에 대응되는 가중치를 찾아 이를 신경망 모형의 출력요소 값으로 활용하였다. 본 연구에서 제안하는 신경망 모형은 임의의 수문 상황에 대해 이에 적합한 CoMOM의 초기 가중치를 결정해 줌으로써 CoMOM과 같은 최적화 모형의 가중치 선정에 따르는 어려움을 극복하는 데 도움이 될 수 있을 것으로 기대된다.

• Journal of Korean Tunnelling and Underground Space Association
• /
• v.9 no.1
• /
• pp.49-62
• /
• 2007

A new earth pressure equation acting on the vertical shafts in cohesionless soils has been proposed by modifying the equations proposed by others. In order to verify the modified equation, model tests which can control uniform wall displacement with depth to radial direction were conducted. Model tests were performed with three different wall friction angles and two different relative densities. The measured values were larger than estimated values when assuming $\lambda=1$ ; smaller than those when assuming $\lambda=1-sin\phi$. The parameter, $\lambda$ is the ratio of tangential stress to vertical stress and is the most critical value in proposed equation. A method which can estimate the earth pressure on vertical shafts in layered soils is also proposed by reasonably assuming the failure surface of layered soils and using the modified equation. In order to verify the proposed method, in-situ measurement data have been collected from the three in-situ vertical shafts installed in layered soils. Most of earth pressures converted from measured data match reasonably well with estimated values using proposed method.

• Journal of Korea Port Economic Association
• /
• v.27 no.3
• /
• pp.135-146
• /
• 2011

A voluminous research on efficiency employs the DEA(Data Envelope Analysis) models. There are, however, only very few that have an interest in the factors influencing such efficiencies. We, furthermore, do not see any studies which analyze the long-term efficiency of the logistics providers using the panel cointegration techniques. The purpose of this paper, hence, is to evaluate the efficiency, analyse its determinants and show a long-term relationship between turnover and the other variables employing the SBM(Slack Based Measure) model, Tobit model, the panel procedure and the FMOLS(Fully Modified OLS). The panel data are composed of 9 individuals and 6 years. The panel cointegrating vectors show that the group coefficient of asset and employees is not only significant but has expected signs, while some of the individual coefficients are insignificant or/and exhibit wrong signs. The panel cointegrating vectors from fully modified OLS also indicate that the estimated coefficients of the panel analysis tend to be overvalued and the asset influences the turnover far greater than the employee does.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2016.05a
• /
• pp.313-316
• /
• 2016

Highly probable cause of the current turmoil critical national infrastructure and private enterprise information systems failures or national event of an accident, it is inevitable huge hit in corporate business as well as a significant impact on the national benefit. This is a trend subject to the stability of the national management information system has a significant impact on people and society, such as banking, telecommunications, transport and energy, which can be extended into a private institution. However, public and private ICT business through the development of Information system Audit or CMMI(Capability Maturity Model Integration) certification check the quality of the target system, but quality check on the reliability of the information system operations after construction is either not promoting met for some safety test results a situation that does not enabled by insufficient. By preventing the disorder or an accident of this study, the diagnostic reliability information systems through inspection and evaluation system development of information systems being established, and to minimize service confusion and study ways to ensure customer - oriented service.

• Composites Research
• /
• v.21 no.5
• /
• pp.15-22
• /
• 2008

This paper describes the results of structural integrity and crashworthiness of Automatic Guideway Transit(AGT) vehicle made of sandwich composites. The applied sandwich composite of vehicle structure was composed of aluminum honeycomb core and WR580/NF4000 glass fabric/epoxy laminate composite facesheet. Material testing was conducted to determine the input parameters for the composite facesheet model, and the effective equivalent damage model fer the orthotropic honeycomb core material. The finite element analysis using ANSYS v11.0 was dont to evaluate structural integrity of AGT vehicle according to JIS E 7105 and ASCE 21-98. Crashworthiness analysis was carried out using explicit finite element code LS-DYNA3D with the lapse of time. The crash condition was frontal accident with speed of 10km/h at rigid wall. The results showed that the structural integrity and crashworthiness of AGT vehicle were proven under the specified loading and crash conditions. Also, the modified Chang-Chang failure criterion was recommended to evaluate the failure modes of composite structures after crashworthiness event.

• Journal of the Korea Society of Computer and Information
• /
• v.29 no.1
• /
• pp.241-249
• /
• 2024

Using the panel data of 31 regions in China from 2006 to 2021, this paper analyzes the efficiency and influencing factors of the fiscal expenditure on compulsory education in China. Through the analysis found that the efficiency of some economically developed regional is lower, but the efficiency of economically backward regional is higher. In order to further analyze the influencing factors of the efficiency of fiscal expenditures on compulsory education, using Tobit model is analyzed. It is found that regional per capita GDP and fiscal self-sufficiency rate have a negative impact on the efficiency of fiscal expenditure on compulsory education. Age structure, percentage of fiscal expenditure on compulsory education and level of teachers have a positive impact on the efficiency of fiscal expenditure on compulsory education. Fiscal decentralization has no significant effect on the efficiency of fiscal expenditure on compulsory education.

• Asia pacific journal of information systems
• /
• v.13 no.4
• /
• pp.229-255
• /
• 2003

This article attempts to examine the well-known law that the increasing returns to scale(IRS) is effective in the Internet business. The effect of IRS is one of the hottest issues in the Internet business sector. Many cases and survey studies support the fact that the IRS phenomenon exists in the Internet business. Executives in Internet business generally give a deep trust on this theory. As the Internet business grows, however, the boundary of the business becomes widened and complicated. And each category of Internet business is characterized with a different business style and economic behavior. It may, therefore, be dangerous to accept that the phenomenon of IRS is applied to all areas of Internet business. For this reason, the research for the close look into the IRS phenomenon should provide significant implications for the managers in the Internet business industry. This article divides the internet business into four sub-areas, and analyzes the IRS phenomenon using AHP/DEA-based full ordering technique. Interpretations are given, based upon the research results, for each sub-area of Internet business, as a guideline of setting business strategies for practical managers.