• 한국컴퓨터정보학회:학술대회논문집
• /
• 한국컴퓨터정보학회 2018년도 제58차 하계학술대회논문집 26권2호
• /
• pp.139-140
• /
• 2018

본 논문에서는 지도 기계 학습 알고리즘 중 하나인 Naive Bayesian (NB) 알고리즘의 데이터 분류 정확도를 향상시키기 위하여 데이터 속성에 Weight를 부여하는 새로운 기법을 제안하였다. 기존에 Decision Tree(DT) 알고리즘의 깊이를 이용하여 Weigth를 부여하는 방법이 제안되었으나, DT를 구축하는데 오버헤드가 크기 때문에 데이터의 실시간 분석이나 자원 제한적인 환경에서의 적용은 어렵다는 단점이 있다. 이를 해결하기 위하여 본 논문에서는 최소한의 데이터를 사용하여 신속하게 DT를 구축하는 Very Fast Decision Tree (VFDT) 알고리즘 기반의 Weight 부여 기법을 제안함으로써 적은 오버헤드로 NB의 정확도를 향상시킨다.

• 한국지능정보시스템학회:학술대회논문집
• /
• 한국지능정보시스템학회 2000년도 추계정기학술대회:지능형기술과 CRM
• /
• pp.439-445
• /
• 2000

다차원의 속성들을 포함한 대용량의 데이터베이스 또는 점보 저장소의 데이터로부터 지식을 추출하고 이를 활용하기 위해서는 데이터 마이닝의 인공지능 기법 중 기계학습을 활용할 수 있다. 본 논문은 질의어를 바탕으로 각 작성들에 가중치를 적용하여 사용자가 원하는 데이터 집합을 분류하고, 사용자 피드백을 통하여 속성 가중치를 동적으로 변화시킴으로써 검색결과를 향상시키는 방법을 제안한다. 본 논문에서는 데이터 집합을 분류해내기 위해서 각 속성간의 거리에 가중치를 적용하는 k-nearest neighbor 분류법을 사용하였고, 속성 가중치를 동적으로 변화시키는 규칙을 추출하기 위한 방법으로는 결정 트리 생성에 의한 규칙(decision rule) 생성 방법을 적용하였다. 검색결과 향상을 \ulcorner이기 위한 실험으로써 온라인 커플매칭(online couple-matching) 시스템의 핵심부문을 구현하고 이를 적용하였다.

• 품질경영학회지
• /
• 제25권2호
• /
• pp.169-188
• /
• 1997

The purpose of this study is concerned with suggesting a new a, pp.oach to evaluating the overall product quality in the sense of the relative efficiency of products, whose quality is measured with 8-dimensional attributes, suggested by Garvin. The attributes included 8 quality measures, that is, performance, features, reliability, conformance, durability, serviceabilit, aesthetics, and perceived quality. This study, also, introduced DEA(Data Envelopment Analysis) as an evaluation tool to tackle the problem of how to measure one product against another when each product can be measured along a number of dimensions, and given that there exists no a priori satisfactory weighting scheme to combine these dimensions into an overall rating for each products. In order to a, pp.y the DEA to evaluating the products, we must define two concepts, such as DMU(Decision Making Units) and input-output relationship. Finally, the suggested method in this study was validated through a case study of the electric jar, and this DEA a, pp.oach can be used in evaluating the other products with multi-attrribute quality.

• 대한인간공학회지
• /
• 제31권1호
• /
• pp.63-76
• /
• 2012

Objective: The main objective of this study is to propose a user-centered assessment metric for ubiquitous services. Background: As the ubiquitous era took off, the interactions between ubiquitous services and users have come to take an important position. It is essential to conceptualize a new assessment model that considers human-system interaction capability with a user-centered design perspective. Method: The evaluation model for the interactivity of ubiquitous service was approached from the concept of usability and inter-personality of services. As a validation study, suggested assessment metric was utilized to evaluate the u-Home service. Priority weighting of each assessment metric was derived using the quantification type-I analysis. Results: To evaluate interactivity, this study suggested a quantitative metric for user testing performed after classifying the interactivity characteristics to contextualization; ubiquity; user experience; and service capability. Conclusion: This study suggest the metric for the ubiquitous service that are experienced in real life, and introduced the concept of ubiquitous service interactivity. Application: The suggested evaluation metric can be used to evaluate interactivity level of ubiquitous service and identify the potential problem and usability requirements at the early stage of service development.

• 디지털산업정보학회논문지
• /
• 제9권3호
• /
• pp.129-144
• /
• 2013

Recognizing a limits on quantitative evaluation of cultural contents' and for its betterment, study aims developing a qualitative evaluation model. For this study, Reception Theory, Semiotics and Psychology were derived for epistemological dimension to contemplate culture contents' essential attribute. To be concrete, cultural contents was examined as experiential products, emotional products, rememberable products and texts. Also, codes of fun, emotion and culture were discussed as intrinsic attributes for cultural contents and how those attributes were expressed or composed in cultural contents was discussed as well. Evaluation items were extracted based on final discussion at the epistemic level, set up the final evaluation model by taking experts' advices on each items. With all those outcomes, qualitative evaluation model for cultural contents was developed. For the importance of each index in the model, priority was granted by weighting on each index. Lastly, evaluation scale was developed for each index. The culture contents' evaluation model developed in study is meaningful not only in drawing qualitative evaluation items of video(image) contents and developing the index and model for the first time, but also its possibility of wide use for other genres.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제11권6호
• /
• pp.2996-3011
• /
• 2017

With rapid growth of web technology and dissemination of smart devices, social networking service(SNS) is widely used. As a result, huge amount of data are generated from SNS such as Twitter, and sentiment analysis of SNS data is very important for various applications and services. In the existing sentiment analysis based on the $Na{\ddot{i}}ve$ Bayes algorithm, a same number of attributes is usually employed to estimate the weight of each class. Moreover, uncountable and meaningless attributes are included. This results in decreased accuracy of sentiment analysis. In this paper two methods are proposed to resolve these issues, which reflect the difference of the number of positive words and negative words in calculating the weights, and eliminate insignificant words in the feature selection step using Multinomial $Na{\ddot{i}}ve$ Bayes(MNB) algorithm. Performance comparison demonstrates that the proposed scheme significantly increases the accuracy compared to the existing Multivariate Bernoulli $Na{\ddot{i}}ve$ Bayes(BNB) algorithm and MNB scheme.

• 한국정보시스템학회지:정보시스템연구
• /
• 제25권2호
• /
• pp.153-171
• /
• 2016

Purpose This study suggests the development of forecasting model for local cable TV advertisement. In order to verify the expected effect of the suggestion, using the causal loop map of System Dynamics, the factors affecting the prospects of cable TV commercial market were divided into 5 groups. Then targeting 97 people involved in the cable TV commercial market in Busan, Ulsan, and Gyeongnam, a survey was conducted on their perception of the current status of local advertisement market and future prospect. Design/methodology/approach The analysis of the collected data shows that workers in advertising and advertisers perceive the influence of cable TV as an advertising media to be high, while clearly understanding the problems of cable TV commercial market. Based on this the effects on the prospects of cable TV commercial market were analyzed and a forecasting method called Weighted Moving Average was applied. In order to improve accuracy of the added value of Weighted Moving Average, the 5 factors were divided into qualitative factors and quantitative factors, and using Multi-attribute Decision Making method, all the factors were normalized and weighting factors were deduced. The result of simulating the prospects of cable TV commercial market using Weighted Moving Average, both qualitative and quantitative factors showed downward turn in the market prospect for the following 10 years. Findings The result reflects generally negative perception of advertisement viewers about the prospects of cable TV commercial market. Compared to the previous studies on domestic cable TV commercials that focused on policy suggestions and surveys on perception of current status, this study has its significance in that it used scientific method and simulation for verification.

• 한국정보통신학회논문지
• /
• 제20권8호
• /
• pp.1581-1586
• /
• 2016

본 논문에서는 산불 발생 시 화재의 확산경로와 속도 예측이 가능한 관제 모니터링 알고리즘을 개발 하였다. 특히 관제 기능에서는 안전 구역 확보를 통해 화재진압에 소요되는 시간을 줄이고 인명, 산림재산 피해를 최소화 하는 방법을 제안한다. 기존 산불 확산 경로 예측 방법에서는 지형, 기상, 연료인자, 영상정보 등을 통해 산불 확산 모델 및 속도를 예측한다. 하지만 이 경우 범위가 넓은 산을 관제하기엔 비용도 많이 소요가 되고, 확산 모델 예측 및 경로 파악에만 집중하여 안전 구역 확보에 대한 노력이 부족한 문제점들이 발생한다. 따라서 본 논문에서는 산불의 이동방향과 속도를 예측하고 화재 진압을 위한 안전구역을 확보하는 알고리즘을 제안한다. 제안된 알고리즘은 온도 변화량 및 연기와 풍향 등의 산불 재난에 따른 속성정보를 분석하여 산불의 이동방향을 예측하고 안전구역을 확보하는 기법이다. 개발된 훈련 관제 모니터링 시스템은 주어진 모의실험 환경에서 산불의 이동 속도 및 이동 방향의 분석이 가능하고 산불에 대한 확산 예측과 진화 및 진압 훈련이 가능한 중앙관제 모니터링 기능을 제공한다.

• 한국산업정보학회논문지
• /
• 제20권4호
• /
• pp.1-9
• /
• 2015

본 연구에서는 다속성 의사결정(MADM)과 게임 이론을 결합하여 무선 자원의 이용효율을 향상시키는 호 수락제어 방법을 제시한다. 다속성 의사결정 방법으로 그레이 관계분석(GRA), 단순부가가중치법(SAW), TOPSIS들을 이용하였는데, 이 방법들은 서로 다른 서비스 품질(QoS)을 갖는 서비스들이 선호하는 대상 네트워크들의 선호도를 계산할 것이다. 이 선호도 값들을 이용한 효용함수를 바탕으로, 사용자가 요구하는 서비스 중에서 서비스 제공자들에게 적합한 서비스를 선택할 수 있도록 비협력적 게임이 진행된다. 요청되는 모든 서비스가 선택될 때까지 게임은 반복적으로 진행되며, 각 단계에서 내쉬균형을 이루는 서비스가 선택되도록 하였다. 서로 다른 특성을 갖는 4개의 무선 랜(WLAN) 시스템 중에서 임의의 2개의 네트워크가 중첩하여 존재하는 경우들을 각각 분석한 결과, 모든 다속성 의사결정 방법들은 서비스 제공자가 얻는 최대 보수의 차이는 있었으나 게임의 각 단계에서 동일한 서비스 선택하는 결과를 얻을 수 있었다.

• The Journal of Asian Finance, Economics and Business
• /
• 제8권3호
• /
• pp.307-318
• /
• 2021

The objective of this study is to analyze the performance of Islamic banks with the Maqashid Shariah approach. The analysis technique used is the Simple Additive Weighting Method (SAW) to solve multi-attribute decision problems. The sampling technique used was purposive sampling while the data came from the annual report of each bank. The results showed that the BTPN Shariah (BTPNS) and Bank Muamalat Indonesia (BMI) are ranked first and second respectively on the Maqashid Shariah Index (MSI) with values of 0.265429 and 0.237110 respectively. Panin Dubai Shariah Bank (PDSB) ranked third with an MSI value of 0.180733, followed by BCA Shariah which ranked fourth with an MSI value of 0.151299. BRI Shariah ranked fifth with an MSI value of 0.128606, followed by BNI Shariah which ranked sixth with an MSI value of 0.124661. Bank Mega Shariah ranked last with an MSI value of 0.087068. Furthermore, there is a relationship (correlation) between ROE, ROA, and OEOI and MSI since each data has a value of 0.000, 0.000, 0.050, and 0.001 respectively, which is smaller than the significance value of 0.05. On the other hand, NPF, TPF, and Asset Growth Rates do not correlate with the MSI since each data has a value of 0.051, 0.252, and 0.215 respectively which is greater than the significance value of 0.05.