• Journal of the Korean Data and Information Science Society
• /
• 제17권1호
• /
• pp.77-86
• /
• 2006

This study is to investigate the error of prediction value with related variables from the recommended system and to examine the error of prediction value with related variables. To decrease the error on the collaborative recommended system on user based, this research explored the effects on the prediction related response pair between raters' demographic variables and Pearson's coefficient and sparsity. The result shows comparative analysis between existing error of prediction value and conditioned one.

• 한국융합학회논문지
• /
• 제12권11호
• /
• pp.91-97
• /
• 2021

본 논문은 불완전한 데이터를 처리하기 위해 본래 규칙개선 문제를 위해 고안되었던 데이터 확장 기법을 사용한다. 이 기법은 사건마다 중요도를 의미하는 가중치를 가질 수 있으며 각 변수를 확률값으로 나타낼 수 있는 특징이 있다. 본 논문에서의 핵심 문제가 손실값과 가장 근사한 확률을 구하여 손실값을 확률로 대치하는 것이므로, 3가지 다른 알고리즘으로 손실값에 대한 확률을 구한 후 이 데이터 구조의 형식으로 저장한다. 그리고 각각의 확률 구조에 대한 평가를 위해 SVM 분류 알고리즘으로 각각의 정보 영역을 분류하는 학습을 한 후, 본래의 정보와 비교하여 얼마나 서로 일치하느냐를 측정한다. 손실값의 대치 확률을 위한 3가지 알고리즘들은 같은 데이터 구조를 사용하고 있으나 접근 방법에서는 서로 다른 특징을 가지고 있어 적용 분야에 따라 다양한 용도로 이용될 수 있기를 기대한다.

• 대한산업공학회지
• /
• 제38권4호
• /
• pp.276-284
• /
• 2012

Classification algorithms generally assume that the data is complete. However, missing values are common in real data sets due to various reasons. In this paper, we propose to use locally linear reconstruction (LLR) for missing value imputation to improve the classification performance when missing values exist. We first investigate how much missing values degenerate the classification performance with regard to various missing ratios. Then, we compare the proposed missing value imputation (LLR) with three well-known single imputation methods over three different classifiers using eight data sets. The experimental results showed that (1) any imputation methods, although some of them are very simple, helped to improve the classification accuracy; (2) among the imputation methods, the proposed LLR imputation was the most effective over all missing ratios, and (3) when the missing ratio is relatively high, LLR was outstanding and its classification accuracy was as high as the classification accuracy derived from the compete data set.

• Journal of Information Technology Applications and Management
• /
• 제23권1호
• /
• pp.97-117
• /
• 2016

Drawing on the theoretical framework of customer value-satisfaction-loyalty, this study examines how scarcity message influences customers' value, satisfaction, and loyalty in the context of group-buying social commerce. Previous studies have argued that scarcity message limiting the avilability of products and service is a source of enhancing customer value. In this regard, this study posits scarcity message as a predictor of customer value. Furthermore, this study classifies customer value into two forms (i.e., utilitarian value and hedonic value) and verfies how scarcity message is associated with them. To test the proposed research model and hypotheses, this study performed structural equation modeling (SEM) analyses, using a total of 292 data collected on users who have experience in purchasing products and service through group-buying social commerce sites such as Coupang, Timon, and WeMakePrice. The key results are as follows : First, scarcity message increases utilitarian and hedonic values and further customer satisfaction. Second, utilitarian value increases customer satisfaction and repurchase intention while hedonic value has nothing to do with them. The findings imply that customers seek to maximize utilitarian value through group-buying social commerce. Finally, this study indicates that repuchase intention depends greatly on customer satisfaction.

• The Journal of Asian Finance, Economics and Business
• /
• 제7권7호
• /
• pp.159-167
• /
• 2020

The study examines the impact of corporate social responsibility (CSR) activity on the firm market value, in particular, market capitalization of tour operators listed on Chinese stock markets. This study employs panel data analysis methods to examine endogeneity concerns in observational data. The balanced panel data includes a total of 1,296 observations with 27 cross-sections of tour operators listed on Chinese stock markets and with 48 time-specific periods from March 2006 to December 2017. The results indicate that CSR activity has a negative impact on the market value of the firm for the concurrent period, but from one-period time lag and afterwards CSR activity has a strong positive impact on the market value and sustains its positive impact on the market value even for a two-period time lag. The findings suggest that the economic effect of CSR activity on the firm market value tends to take some degree of lagged effects to be fully showcased in the market capitalization of tour operators and travel companies listed on Chinese stock markets. The findings suggest that, though CSR activity may carry some financial risk for an immediate short-term, tour operators must put a lot of time and effort into making CSR actions effective.

• Journal of applied mathematics & informatics
• /
• 제29권3_4호
• /
• pp.713-719
• /
• 2011

The problem of approximation from a set of scattered data arises in a wide range of applied mathematics and scientific applications. In this study, we present a quasi-interpolatory approximation scheme for scattered data approximation problem, which reproduces a certain space of polynomials. The proposed scheme is local in the sense that for an evaluation point, the contribution of a data value to the approximating value is decreasing rapidly as the distance between two data points is increasing.

• 유통과학연구
• /
• 제15권3호
• /
• pp.49-60
• /
• 2017

Purpose - The aims of this study are follows. We investigated to find out how country image and brand image affect the consumer perceived value, consumer attitude, and purchase intention of foreign infant foods in China. Especially, we focused on investigate for the moderating role of consumer knowledge between national image, brand image and consumer perceived value of foreign infant foods in China. Research design, data, and methodology - This study analyzed the effect of national image and brand image on purchase intention through consumer perceived value and consumer attitude. This study collected data for empirical analysis of Chinese consumers who have been purchase experience infant foods in China. 256 copies of questionnaire data were used for substantial analysis. Before testing the hypothesis, factor analysis was conducted to test the construct validity of measurement items. Hypotheses about effects between variables were verified using structural equation modeling analysis and hierarchical regression analysis. Results - First, the country image had a positive effect on consumer perceived value of foreign infant foods. Second, the brand image had a positive effect on consumer perceived value of foreign infant foods. Third, the consumer perceived value had a positive effect on consumer attitude. Fourth, the consumer attitude had a positive effect on purchase intention. Fifth, the consumer knowledge was moderating roles between brand image and consumer perceived value of foreign infant foods. However, the consumer knowledge did not effect of moderating between country image and perceived value of consumers. Conclusions - First, the impact of country image and brand image on consumer perceived value of foreign infant foods in China can be seen as a universal psychology of consumers who trust pure foreign products such as high quality, technology, etc. Second, consumer perceived value of foreign infant foods has a positive effect on consumer attitude, and this attitude is leading to purchase intention. Third, the consumer knowledge between brand image and perceived value acts as a moderating variable. It means that the consumer's knowledge can shape the perception of the brand image more strongly.

• 한국의류산업학회지
• /
• 제13권1호
• /
• pp.25-31
• /
• 2011

The purpose of this study was to investigate the influence of luxury brand value on luxury brand preferences according to chemyon(social-face). For data collection, research questionnaires were responded by 222 females over 20 years old consumers living in Pusan and Ulsan. The collected data were analyzed using SPSS 14.0 for windows Package by frequency analysis, the factor analysis using Varimax, reliability analysis, multi-regression analysis, and chow-test. The results of this study were as follows: First, concerning the worth of luxury brands that qualify value, conspicuous value, unique value, and emotional value significantly affected the consumers' luxury brand preferences. On the other hand, only social value of the luxury brand did not have a significant effect on the consumers luxury brand preferences. Second, the study was to investigate whether the luxury brand value has an effect on the brand preferences according the consumer's chemyon or not. The results showed different effects of perceived values on luxury brand preferences between the low chemyon group and the high chemyon group. The high chemyon group had an effect on conspicuous value, social value, and emotional value, while the low chemyon group had an effect on quality value for luxury brand preference.

• Journal of the Korean Data and Information Science Society
• /
• 제21권1호
• /
• pp.185-194
• /
• 2010

In recommender systems which are used widely at e-commerce, collaborative filtering needs the information of user-ratings and neighbor user-ratings. These are an important value for recommendation in recommender systems. We investigate the in-formation of rating in NBCFA (neighbor Based Collaborative Filtering Algorithm), we suggest new algorithm that improve prediction accuracy of recommender system. After we analyze relations between two variable and Error Value (EV), we suggest new algorithm and apply it to fitted line. This fitted line uses Least Squares Method (LSM) in Exploratory Data Analysis (EDA). To compute the prediction value of new algorithm, the fitted line is applied to experimental data with fitted function. In order to confirm prediction accuracy of new algorithm, we applied new algorithm to increased sparsity data and total data. As a result of study, the prediction accuracy of recommender system in the new algorithm was more improved than current algorithm.

• 지능정보연구
• /
• 제25권2호
• /
• pp.99-122
• /
• 2019

IDC(International Data Corporation) 사(社)의 최근 보고서에 따르면, 2025년에는 2016년에 생성된 데이터의 10배에 달하는 163제타바이트의 데이터가 생성될 것이고 그 주체의 비중은 소비자에서 기업으로 이동하고 있다고 한다. 이러한 소위 '빅데이터의 물결'은 도래하고 있고 그 파장은 산업 전반적으로 영향을 미칠 것이다. 따라서, 방대한 데이터를 효과적으로 관리하는 것은 기업의 관점에서 그 어느 때보다 더 중요하다. 하지만, IT 투자에 대한 효과를 측정한 선행 연구는 다수 존재함에도 불구하고 빅데이터 투자 효과를 측정한 선행 연구는 거의 전무한 실정이다. 따라서, 해당 투자 효과를 정량적으로 분석한다면 기업의 의사 결정을 도울 수 있을 것이다. 본 연구는 효율적 시장 가설을 이론적 바탕으로 둔 사건연구방법론(Event Study Methodology)을 적용하여, 기업의 빅데이터 투자가 시장 투자자들의 반응에 미치는 영향을 측정하였다. 또한, 보다 심층적으로 이 효과를 분석하기 위해서 5가지 하위 변수를 설정했고 그 내용은 기업 크기 구분, 산업 구분(Finance와 ICT), 투자 구축 완료 구분, 벤더 유무 구분이다. 분석 결과, 91개 기업은 빅데이터 투자 공시 이후 시장 가치가 평균 0.92% 상승한다는 사실을 확인하였다. 특히 Finance 기업, non-ICT 기업, 시가 총액이 작은 기업, 빅데이터 전문 벤더 기업을 통해 투자한 기업, 그리고 빅데이터 시스템이 구축 완료됐다는 공시에 해당하는 기업의 시장 가치가 두드러지게 상승한다는 사실을 알 수 있었다. 본 연구는 빅데이터 투자 효과를 측정한 선행 연구가 거의 전무하다는 점에서 학문적인 의의를 지니고, 빅데이터 투자를 고려 중인 기업 의사 결정자들에게 실질적인 참고 자료가 될 수 있다는 점에서 실무적인 시사점을 갖는다.