Earticle

최근 인터넷 등 정보통신 기술의 발달로 인해 언제 어디서나 인터넷을 이용할 수 있는 환경이 구축 되었으며, 이로인한 사이버위협은 다양한 경로를 통해 시도되고 있다. 본 연구에서는 사이버위협 중 지속적으로 증가 추세인 DDoS 예측 모델링하기 위해 이벤트 데이터를 근거로 하여 통계적 기법을 통해 DDoS 위험지수 예측식을 도출하였고, 도출된위험지수를 정량화하였다. 제시된 위험지수를 활용하여 DDoS 위협에 대해 사전 대응정책을 세움으로써 피해를 최소화시킬 수 있는 객관적이고 효율적인 예측 모델이 될 것으로 기대한다.

With the development of information and communication technologies like internet, the environment where people are able to access internet at any time and at any place has been established. As a result, cyber threats have been tried through various routes. Of cyber threats, DDoS is on the constant rise. For DDoS prediction modeling, this study drew a DDoS security index prediction formula on the basis of event data by using a statistical technique, and quantified the drawn security index. It is expected that by using the proposed security index and coming up with a countermeasure against DDoS threats, it is possible to minimize damage and thereby the prediction model will become objective and efficient.

KCB는 2017년 공공 빅데이터 표준분석모델 수행기관으로 선정되어, 표준 체납자 예측 모형을 개발하고 기존 체납 정보 시스템을 개선하였다. 해당 과제는 국가행정의 기반이 되는 세금의 비효율적인 관리 및 징수 프로세스와 시스템을 개선하고, 데이터 기반의 과학적 국정운영 및 의사결정을 도모하였다는 평가를 받았다. 본 고에서는 체납자 회수 가능성 예측 모형의 주요 내용과 시뮬레이션 분석을 통한 기대효과를 살펴보고자 한다. 첫째, 체납자의 신용정보와 체납 정보의 데이터 융합 및 탐색적 자료 분석(EDA). 둘째, 로지스틱 회귀분석과 세 가지 머신 러닝 방법론(Recursive Partitioning, Neural Network, Random Forest) 의 예측 성능 비교를 통한 모형개발과정. 셋째, 예측 결과의 시스템 적용 및 시사점. 끝으로, 예측 시뮬레이션 분석을 통한 모형 적용 시 세금 조기 환수 및 비용절감 기대효과.

This study was conducted to develop the fermentation kinetic model for the prediction of acidity and pH changes in Kimchi as a function of fermentation temperatures. The fitness of the model was evaluated using traditional two-step method and an alternative non-linear regression method. The changes in acidity and pH during fermentation followed the pattern of the first order reaction of a two-step method. As the fermentation temperature increased from 4$^{\circ}C$ to 28, the reaction rates of acidity and pH were increased 8.4 and 7.6 times, respectively. The activation energies of acidity and pH were 16.125 and 16.003kcal/mole. The average activation energies of acidity and pH using a non-linear method were 16.006 by the first order and 15.813 kcal/mole by the zero order, respectively. The non-linear procedure had better fitting 개 experimental data of the acidity and pH than two-step method. The shelf-lives based on the time to reach the 1.0% of acidity were 33.1day at 4$^{\circ}C$ and 2.8 day 28$^{\circ}C$.

In this paper, the contact force between two colliding bodies is modeled by using Hertz's force-displacement law and nonlinear damping function. In order to verify the appropriateness of the proposed contact force model, the drop type impact test is carried out for different impact velocities and different materials of the impacting body, such as rubber, plastic and steel. In the drop type impact experiment, six photo interrupters in series close to the collision location are installed to measure the velocity before impact more accurately. The characteristics of contact force model are investigated through experiments. The parameters of the contact force model are estimated using the optimization technique. Finally the estimated parameters are used to predict the impact force between two colliding bodies in opening action of the magnetic contactor, a kind of switch mechanism for switching electric circuits.

In the default prediction problem, the cost from the failure of forecasting defaults is much bigger than that of forecasting non-defaults. The cost asymmetry is deeper in the corporate default prediction than the retail as corporate loan portfolios are not granular. However, the two types of costs are treated equally in general as default prediction models are usually estimated to minimize prediction errors or maximize statistical performance. This practice might not fulfill the goal of risk management to minimize economic losses. To mitigate this issue, this study apply cost-sensitive learning approach to default prediction, which minimizes economic costs instead of statistical errors. We define economic costs and test them for various levels of the cost asymmetry by employing Logistic regression, XGBoost, and LightGBM. As a result of empirical experiments with Taiwanese and Polish corporate default data, we first find that the proposed cost-sensitive models are superior to the cost-insensitive counterparts in terms of economic cost, mostly regardless of the cost asymmetry scenarios. Secondly, nevertheless, the decreases in the statistical performance are relatively small – economic costs decrease 24.6% at the expense of the decrease in AUC of 4.6% on average. This suggests that financial firms can adopt the proposed default prediction models without violating the regulatory requirement on model quality. Lastly, we find that the features of high prediction power in the cost-sensitive and insensitive models are different, which has an important implication for credit monitoring.

In the default prediction problem, thecost from thefailure of forecasting defaults is much biggerthan that of forecastingnon-defaults. Thecost asymmetryis deeperin the corporatedefaultpredictionthan theretailas corporateloan portfolios arenot granular. However, thetwo typesof costs aretreatedequallyin generalas defaultpredictionmodels areusually estimatedto minimizepredictionerrorsor maximizestatistical performance. This practicemight not fulfill thegoal of risk managementto minimizeeconomiclosses. To mitigatethis issue, this study apply cost-sensitivelearningapproach to defaultprediction, which minimizeseconomiccosts insteadof statistical errors. Wedefineeconomic costs and testthemfor various levelsof thecost asymmetryby employingLogistic regression, XGBoost, and LightGBM. As a resultof empiricalexperimentswith Taiwaneseand Polish corporate default data, we first find that the proposed cost-sensitivemodels are superiorto thecost-insensitivecounterpartsin termsof economiccost, mostly regardless of thecost asymmetryscenarios. Secondly, neverthele,ssthedecreasesin thestatistical performanceare relativelysmall – economic costs decrease24.6% at theexpenseof the decrease in AUC of 4.6% on average. This suggests that financial firms can adopt the proposed default prediction models without violating the regulatory requirement on model quality. Lastly, we find that the features of high prediction power in the cost-sensitive and insensitive models are different, which has an important implication for credit monitoring.

Bankruptcy prediction has been studied to develop predictive models based on financial variables. Using only financial variables may be insufficient in bankruptcy prediction modeling because they do not reflect the latest information, essentially when using past corporate accounting information. Thus, exploiting qualitative information with quantitative information is required to supplement the limited accounting information. Among big data analytics techniques, text mining is used for processing qualitative information. In this study, we propose an integrated approach for bankruptcy prediction using market sentiment extracted from economic news as qualitative information and financial variables as quantitative information for bankruptcy prediction. Unlike previous sentiment analysis approaches, consideration of topics extracted from economic news in sentiment analysis is included to mitigate the ambiguity of capturing the sentiment for single terms. This study validates the effectiveness of incorporating topic-based market sentiment into the conventional bankruptcy prediction model using financial variables in terms of predictive performance.

본 연구는 건설 현장의 CCTV를 활용한 실시간 침수 예측 시스템을 개발하는 것을 목표로 한다. CCTV로부터 침수 깊이를 추정하기 위한 모델을 구축하고, 추정된 침수 깊이를 관측 데이터와 비교하 여 성능을 검증하였다. 분석 결과 높은 상관관계(R2 = 0.99)와 약 3cm의 RMSE를 보여 정량적 침수 모니터링에 CCTV를 활용할 수 있는 가능성을 확인했다. 또한, 취약 지역에 대한 실시간 침수 예측 방법을 제안했으며 XP SWMM 시뮬레이션에서 도출된 침수 특성을 활용하여 ANN(인공 신경망)과 CNN(합성곱 신경망)을 기반으로 한 모델을 학습시켰다. 모델의 성능은 평균 절대 백분율 오차 (MAPE)를 사용하여 평가되었으며, 전체 평균 오차율은 침수면적 예측에서 8.89%, 그리드 기반 침수 깊이 예측에서 19.49%가 나타났다. 향후 연구는 실시간 CCTV 침수 모니터링과 AI 모델을 통합하여 예측 정확도를 높일 수 있을 것이라 판단된다.

This study aims to develop a real-time inundation prediction system using CCTV at construction sites. A model was established to estimate inundation depth from CCTV, and its performance was validated by comparing the estimated inundation depth with observed data. The results demonstrated a high correlation (R2 = 0.99) and an RMSE of approximately 3 cm, confirming the feasibility of using CCTV for quantitative inundation monitoring. Furthermore, a real-time inundation prediction method for vulnerable areas was proposed. inundation characteristics derived from XP SWMM simulations were used to train a model based on ANN (Artificial Neural Networks) and CNN (Convolutional Neural Networks). The model's performance was evaluated using the Mean Absolute Percentage Error (MAPE), with overall average error rates of 8.89% for inundation area predictions and 19.49% for grid-based inundation depth predictions. Future efforts will focus on integrating real-time CCTV inundation monitoring with the AI model to enhance its predictive accuracy.

The battery market is experiencing rapid growth due to advancements in technology and increased recycling efforts. Verifying the suitability of developed batteries through rate capability experiments, which measure capacity based on charging and discharging speeds, is essential but resource-intensive and time-consuming. This research proposes a method to predict battery rate capability using a polynomial regression model based on similar data groups, aiming to shorten these experiments. The research was conducted in two main stages, namely the construction of the dataset and the development of the predictive model. Data was collected from experimental graphs in existing literature and new experiments on Coin Cell batteries. Through preprocessing steps including deduplication, interpolation, and extrapolation, a comprehensive dataset was created. A combined Quadratic and Linear Piecewise Interpolation method was developed to handle missing data efficiently. In the model development stage, polynomial regression models were created for groups of similar battery data, allowing accurate predictions for partial rate capability experiments. Experimental results demonstrated high accuracy, significantly reducing the need for extensive testing. The proposed method offers substantial time and resource savings, enhancing the efficiency of the battery development process.

Protein phosphatase manganese dependent 1D (PPM1D) is one of the Ser/Thr protein phosphatases belongs to the PP2C family. They play an important role in cancer tumorigenesis of various tumors including neuroblastoma, pancreatic adenocarcinoma, medulloblastoma, breast cancer, prostate cancer and ovarian cancer. Even though PPM1D is involved in the pathophysiology of various tumors, the three dimensional protein structure is still unknown. Hence in the present study, homology modelling of PPM1D was performed. 20 different models were modelled using single- and multipletemplate based homology modelling and validated using different techniques. Best models were selected based on the validation. Three models were selected and found to have similar structures. The predicted models may be useful as a tool in studying the pathophysiological role of PPM1D.

With the rapid development of social media, users' next location prediction has become an important research direction, which can provide personalized travel suggestions for users. However, existing methods ignore the semantic relationship between users' historical and current trajectories. This paper proposes a new method for predicting the user's next location to solve this problem. We first process the user POI data as trajectory data, use the attention mechanism to extract similar features of the user's historical trajectories, and then combine them with the current trajectory features to obtain the user's next location recommendation. The experimental results show that our proposed model performs satisfactorily on a real dataset.

Accurate prediction of breakdown voltage (BDV) in cryogenic environments is critical for ensuring the operational safety and reliability of high-temperature superconducting (HTS) systems. In such systems, dielectric failure due to extreme thermal and electrical stresses can lead to catastrophic malfunction. This study presents a second-order polynomial regression model that quantitatively predicts BDV in liquid nitrogen (LN₂) as a function of electrode gap and diameter. The model was developed based on experimentally measured data and incorporates nonlinear and interaction effects between geometric variables. Statistical validation confirmed its high predictive accuracy (R² = 0.9917), demonstrating robustness. This modeling approach enables pre-operational insulation design optimization and may be embedded into digital twin frameworks for real-time diagnostics. The findings offer both theoretical insights and practical tools for the development and deployment of next-generation cryogenic insulation systems in HTS applications.

Bradykinin receptor B2, a GPCR protein, binds with the inflammatory mediator hormone bradkynin. It plays an important role in cross-talk between the renin-angiotensin system (RAS) and the kinin-kallikrein system (KKS). Also, it is involved in many processes including vasodilation, edema, smooth muscle spasm and pain fiber stimulation. Hence, studuying the structural features of the receptor becomes important. But the unavailability of the three dimensional structure of the protein makes the analysis difficult. Hence we have performed the homology modelling of Bradykinin receptor B2 with 5 different templates. 25 different homology models were constructed. Two best models were selected based on the model validation. The developed models could be helpful in analysing the structural features of Bradykinin receptor B2 and in pathophysiology of various disorders related to them.

Thromboxane A2 receptors (TXA2-R) are the G protein coupled receptors localized on cell membranes and intracellular structures and play pathophysiological role in various thrombosis/hemostasis, modulation of the immune response, acute myocardial infarction, inflammatory lung disease, hypertension and nephrotic disease. TXA2 receptor antagonists have been evaluated as potential therapeutic agents for asthma, thrombosis and hypertension. The role of TXA2 in wide spectrum of diseases makes this as an important drug target. Hence in the present study, homology modeling of TXA2 receptor was performed using the crystal structure of squid rhodopsin and night blindness causing G90D rhodopsin. 20 models were generated using single and multiple templates based approaches and the best model was selected based on the validation result. We found that multiple template based approach have given better accuracy. The generated structures can be used in future for further binding site and docking analysis.

Purpose: This study was conducted to construct and test a structural model on family life satisfaction of aged individuals living at home. The conceptual model was based on Bandura's self-efficacy and social cognitive theories (1977; 1986) and Bowen's (1976) family systems theory. Methods: From January 25 to March 5, 2016, 227 older adults living at home completed a structured questionnaire. Data were analyzed to calculate the direct and indirect effects of factors affecting family life satisfaction. SPSS WIN 20.0 and AMOS 20.0 were used. Results: The hypothetical model was a good fit for the data. The model fit indices were ${\chi}^2=78.05$, ${\chi}^2/df=1.35$, RMSR=.02, GFI=.98, AGFI=.96, NFI=.94, CFI=.98, and RMSEA=.05. Family life satisfaction was positively affected by perceived collective family efficacy, status of physical health, family communication, and family support. Depression resulted in a significant negative effect. Family differentiation had a significant indirect effect on family life satisfaction. The model explained 76% of variance in family life satisfaction. Conclusion: Perceived collective family efficacy, status of physical health, depression, family differentiation, family communication, and family support were significant factors explaining family life satisfaction among older adults staying at home. Further research should be conducted to seek intervention strategies to improve family life satisfaction among older adults living at home by focusing on the respective contributing factors.

In the metal-cutting industry, energy prediction is important for environmentally-conscious manufacturing because it enables a numerical anticipation of the energy consumption from the input of the process parameters, and therefore it contributes to the increasing of the energy-efficiency of the machine-tool operations. This paper proposes an energy-prediction modeling approach for numerical-control programs based on historical machine-monitoring data that have been collected from machine-tool operations. The proposed approach can create accurate energy-prediction models that forecast the energy that is consumed by the execution of a numerical-control program. Also, it can create machine-specific energy-prediction models that accommodate the variety of shop-floor machining contexts. For this purpose, it uses MTConnect to represent the machine-monitoring data to embody an interoperable data-collection environment regarding the shop floor. This paper also presents a case study to show the feasibility and practicability of the proposed approach.

The purpose of this study is to develop soil salt prediction model for the estimation of irrigation water requirements for dry field crops in reclaimed tidelands. The simulation model based on water balance equation, salt balance equation, and salt storage equation was developed for daily prediction of sa]t concentration in root zone. The data obtained from field measurement during the growing period of tomato were used to evaluate the applicability of this model. The results of this study are summarized as follows: 1.The optimum irrigation point which maximizes the crop yield in reclaimed tidelands of silt loam soil while maintaining the salt concentration within the tolerance level, ws found to be pF 1.6, and total irrigation requirement after transplanting was 602mm(6.7 mm/day)for tomato. 2.When the irrigation point was pF 1.6, the deviation between predicted and measured salt concentration was less than 4 % at the significance level of 1 7% 3.Since the deviations between predicted and measured values data decrease as the amount of irrigation water increases, the proposed model appear to be more suitable for use in reclaimed tidelands. 4.The amount of irrigation water estimated by the simulation model was 7.2mm/day in the average for cultivating tomato at the optimum irrigation point of pF 1.6.The simulation model proposed in this study can be generalized by applying it to other crops. This, model, also, could be further improved and extended to estimate desalinization effects in reclaimed tidelands by including meteorological effect, capillary phenomenon, and infiltration.

This paper introduces a new framework of predicting the arrival time of an aircraft by incorporating the probabilistic information of what type of trajectory pattern will be applied by human air traffic controllers. The proposed method is based on identifying the major patterns of vectored trajectories and finding the statistical relationship of those patterns with various traffic complexity factors. The proposed method is applied to the traffic scenarios in real operations to demonstrate its performances.