• 지능정보연구
• /
• 제25권2호
• /
• pp.167-194
• /
• 2019

본 논문은 키워드검색광고와 관련하여 의사결정자인 광고주의 입장에서 분석한 통계모델 기반 검색엔진최적화(Search Engine Optimization)논문이다. 일반적으로 키워드입찰은 노출순위를 대상으로 하는 입찰가액에 의해 이루어지고 있다. 그런데, 대부분 광고주는 수천 개 이상의 많은 키워드를 관리함에 있어, 매시간적으로 바뀌는 키워드별 입찰가액을 통해 입찰광고시스템을 관리하고 있는데, 사실상 시간과 인력자원측면에서 비효율적이다. 따라서, 본 논문에서는 기존의 입찰가액을 중심으로 하는 입찰시스템에 대해 의문점을 제기하고, 새로운 관점에서 노출순위를 의사결정변수로 하는 새로운 검색광고모델을 재정의하여 제시하였다. 새로운 검색광고모델에 대한 최적화실증분석을 위해 예측모델과 최적화모델을 제시하였다. 연구과정은 우선 키워드의 특성에 따라 키워드그룹을 원천 제조브랜드 유통브랜드의 범주화기준을 제시한 후, PC 와 모바일 매체별로 대표 키워드 선정한 후 노출순위와 클릭률이 비선형분포임을 보였고, 통계적 관계를 검토하였다. 클릭률예측 및 입찰가액예측을 위한 통계적 시나리오를 제시하였고, 적합성 분석을 통해 최적의 예측모델을 선정한 후, 선정된 예측모델을 기반으로 하여 클릭률과 기대이익(전환율)에 관한 최적화목적함수를 정의하고 실증분석을 진행하였다. 분석결과, 본 논문에서 제시한 검색광고모델은 클릭률 기반의 클릭수와 전환율 기반의 기대이익으로 표현되는 최적화모델 모두에서 개선효과가 있음을 확인하였다. 다만, 기대이익 최적화모델의 경우에는 핵심키워드임에도 불구하고 기대이익이 낮아 광고에서 배제되는 문제를 있음을 확인하고 대안을 제시했다. 마코브체인분석을 통해 핵심 경유키워드 개념을 도입하였고, 최적화목적함수에 대해 핵심경유키워드의 기회이익을 반영한 최적화수정모델을 제시하여 적용가능성을 확인하였다. 본 논문은 키워드입찰시스템의 의사결정변수를 노출순위의 관점으로 전환하는 새로운 모델을 제안하였고, 키워드 범주별 및 노출순위 기반의 통계적 예측을 제시하고, 포트폴리오 구성에서의 최적화실증분석을 통해 노출순위 기반 예측모델의 유효성을 확인함과 동시에, 키워드간의 확산효과를 포함하는 수정모델제시 등 전략적인 입찰을 제안한 점에 시사점이 있다.

• 한국HCI학회:학술대회논문집
• /
• 한국HCI학회 2006년도 학술대회 1부
• /
• pp.970-975
• /
• 2006

There is a tradeoff between user's privacy and utility of context-aware services in ubiquitous computing environments. Many privacy models have been proposed to support the disclosure of personal information at different levels of detail, in ubiquitous computing environments. However, most of these models do not allow for explicit criteria to assess the benefit users are likely to reap by disclosing their personal information. In this paper, we propose an automated decision making mechanism that evaluates the "benefit of disclosure" for the users based on trust relationships between users and information requesters and manages the disclosure of user's personal information accordingly. Unlike other trust models, we do not regard the reputation of an information requester as sufficient to determine his/her trustworthiness. Instead, we represent trustworthiness as a function of information requester's reputation in the eyes of the user and his/her competence in a given context. To validate our mechanism, we apply it to context-aware healthcare application that monitors physiological condition of a user.

• Coupled systems mechanics
• /
• 제8권6호
• /
• pp.489-500
• /
• 2019

With drinking water standards becoming more rigorous and increasing demands for additional water quantities, while water resources are becoming more polluted, mathematical models became an important tool to improve water treatment processes performance in the water supply system. Water treatment processes models reflect the knowledge of the processes and they are useful tools for water treatment process optimization, design, operator training for decision making and fundamental research. Unfortunately, in the current practice of drinking-water production and distribution, water treatment processes modeling is not successfully applied. This article presents a review of some existing water treatment processes simulators and the experience of their application and indicating the main weak points of each process. Also, new approaches in the modeling of water treatment are presented and recommendations are given for the work in the future.

• Nuclear Engineering and Technology
• /
• 제50권4호
• /
• pp.532-541
• /
• 2018

Decision support systems are a key focus of research on developing control rooms to aid operators in making reliable decisions and reducing incidents caused by human errors. For this purpose, models of complex systems can be developed to diagnose causes or consequences for specific alarms. Models applied in safety systems of complex and safety-critical systems require rigorous and reliable model building and testing. Multilevel flow modeling is a qualitative and discrete method for diagnosing faults and has previously only been validated by subjective and qualitative means. To ensure reliability during operation, this work aims to synthesize a procedure to measure model performance according to diagnostic requirements. A simple procedure is proposed for validating and evaluating the concept of multilevel flow modeling. For this purpose, expert statements, dynamic process simulations, and pilot plant experiments are used for validation of simple multilevel flow modeling models of a hydrocyclone unit for oil removal from produced water.

• 한국문헌정보학회지
• /
• 제31권1호
• /
• pp.185-213
• /
• 1997

본 연구는 정보기술환경의 변화에 따라 대학의 정보관리조직구조를 CIO 조직체제로 변모해 온 과정과 새로운 CIO 조직구조 속에서 대학도서관과 컴퓨터센터의 상호협력관계를 구명하여 이를 국내대학에 적용시킬 수 있는 방안을 제시하였다. 주로 미국대학에서 활용하고 있는 다양한 CIO 조직 모형들의 특성과 이 모형에서 도서관과 컴퓨터센터와의 상호협력관계에 관한 실증적 분석이 이루어졌다.

• 대한산업공학회지
• /
• 제18권1호
• /
• pp.105-120
• /
• 1992

An expert system was developed as a framework of integrating diverse and multifactored ergonomic knowledge to investigate its effectiveness in ergonomic workplace design and evolution. Although numerous computer-assisted approaches have been made to overcome the lack of integrated design principles, those models being used require very specific information of various design activities that may not be available in the design stage. On the other hand, an expert system would be an effective design aid that is capable of guiding the designer to solve a problem. However, most expert systems lack detailed evaluation capabilities due to a qualitative nature of inference mechanisms. Furthermore, those approaches were independently developed, focusing mostly on a single aspect such as biomechanics, physiology, etc. In this paper, a design framework was developed which takes advantage of expert system metholologies, a relational data base and existing ergonomic models. The pattern-directed, rule-based expert system allows the designer to gradually formulate and subsequently evaluate workplace design. A comprehensive and modularized knowledge base was built incorporating biomechanics, physiology and psychophysics, which is, in turn, capable of accessing not only qualitative knowledge but complex analytic evaluation models and massive information in the data base through an interface. A conflict resolution strategy using multiple criteria decision-making schemes was also employed to reconcile multiple design alternatives.

• Journal of Obesity & Metabolic Syndrome
• /
• 제27권4호
• /
• pp.213-222
• /
• 2018

Bariatric surgery has evolved from a surgical measure for treating morbid obesity to an epochal remedy for treating metabolic syndrome as a whole, which is represented by type 2 diabetes mellitus. Numerous clinical trials have advocated bariatric or metabolic surgery over nonsurgical interventions because of markedly superior metabolic outcomes in morbidly obese patients who satisfy traditional criteria for bariatric surgery (body mass index [BMI] >$35kg/m^2$) and in less obese or simply overweight patients. Nevertheless, not all diabetes patients achieve the most desirable outcomes; i.e., diabetes remission after metabolic surgery. Thus, candidates for metabolic surgery should be carefully selected based on comprehensive preoperative assessments of the risk-benefit ratio. Predictors for diabetes remission after metabolic surgery may be classified into two groups based on mechanism of action. The first is indices for preserved pancreatic beta-cell function, including younger age, shorter duration of diabetes, and higher C-peptide level. The second is the potential for an insulin resistance reduction, including higher baseline BMI and visceral fat area. Several prediction models for diabetes remission have been suggested by merging these two to guide the joint decision-making process between clinicians and patients. Three such models, DiaRem, ABCD, and individualized metabolic surgery scores, provide an intuitive scoring system and have been validated in an independent external cohort and can be utilized in routine clinical practice. These prediction models need further validation in various ethnicities to ensure universal applicability.

• Journal of Gastric Cancer
• /
• 제19권3호
• /
• pp.235-253
• /
• 2019

Gastric cancer (GC) is one of the deadliest malignancies in the world. Currently, clinical treatment decisions are mostly made based on the extent of the tumor and its anatomy, such as tumor-node-metastasis staging. Recent advances in genome-wide molecular technology have enabled delineation of the molecular characteristics of GC. Based on this, efforts have been made to classify GC into molecular subtypes with distinct prognosis and therapeutic response. Simplified algorithms based on protein and RNA expressions have been proposed to reproduce the GC classification in the clinical field. Furthermore, a recent study established a single patient classifier (SPC) predicting the prognosis and chemotherapy response of resectable GC patients based on a 4-gene real-time polymerase chain reaction assay. GC patient stratification according to SPC will enable personalized therapeutic strategies in adjuvant settings. At the same time, patient-derived xenografts and patient-derived organoids are now emerging as novel preclinical models for the treatment of GC. These models recapitulate the complex features of the primary tumor, which is expected to facilitate both drug development and clinical therapeutic decision making. An integrated approach applying molecular patient stratification and patient-derived models in the clinical realm is considered a turning point in precision medicine in GC.

• 시스템엔지니어링학술지
• /
• 제16권2호
• /
• pp.67-77
• /
• 2020

Traditionally nuclear thermal hydraulic and nuclear safety has relied on numerical simulations to predict the system response of a nuclear power plant either under normal operation or accident condition. However, this approach may sometimes be rather time consuming particularly for design and optimization problems. To expedite the decision-making process data-driven models can be used to deduce the statistical relationships between inputs and outputs rather than solving physics-based models. Compared to the traditional approach, data driven models can provide a fast and cost-effective framework to predict the behavior of highly complex and non-linear systems where otherwise great computational efforts would be required. The objective of this work is to develop an AI algorithm to predict the peak fuel cladding temperature as a metric for the successful implementation of FLEX strategies under extended station black out. To achieve this, the model requires to be conditioned using pre-existing database created using the thermal-hydraulic analysis code, MARS-KS. In the development stage, the model hyper-parameters are tuned and optimized using the talos tool.

• Journal of Information Processing Systems
• /
• 제18권2호
• /
• pp.208-222
• /
• 2022

Through the analysis of the coordination mechanism of the supply chain system of China's automobile manufacturing industry, the factors affecting the supply subsystem, the manufacturing subsystem, the sales subsystem, and the consumption subsystem are sorted out, the supply chain coordination index system based on the influence factor of four subsystems is established. The evaluation models of the coordination degree in the subsystem of the supply chain, the coordination degree among the subsystems, and the comprehensive coordination degree are established by using the efficiency coefficient method and the collaborative entropy method. Experimental results verify the accuracy of the evaluation model using the empirical analysis of the collaborative evaluation index data of China's automobile manufacturing industry from 2000 to 2019. The supply chain synergy of automobile manufacturing industry was low from 2001 to 2005, and it increased to a certain extent from 2006 to 2008 with a small growth rate from 0.10 to 0.15. From 2009 to 2013, the supply chain synergy of automobile manufacturing industry increased rapidly from 0.24 to 0.49, and it also increased rapidly but fluctuated from 2014 to 2019, first rising from 0.68 to 0.84 then dropping to 0.71. These results provide reference for the development of China's automobile manufacturing supply chain system and scientific decision-making basis for the formulation of relevant policies of the automobile manufacturing industry.