Analysis of Incomplete Data with Nonignorable Missing Values

김현정,Kim, Hyun-Jeong The Korean Data and Information Science Society 2002 한국데이터정보과학회지 Vol.13 No.2

In the case of "nonignorable missing data", it is necessary to assume a model dealing with the missing on each situations. In this article, for example, we sometimes meet situations where data set are income amounts in a survey of individuals and assume a model as the values are the larger, a missing data probability is the higher. The method is to maximize using the EM(Expectation and Maximization) algorithm based on the (missing data) mechanism that creates missing data of the case of exponential distribution. The method started from any initial values, and converged in a few iterations. We changed the missing data probability and the artificial data size to show the estimated accuracy. Then we discuss the properties of estimates.

Comparison study of SARIMA and ARGO models for in influenza epidemics prediction

Jung, Jihoon,Lee, Sangyeol The Korean Data and Information Science Society 2016 한국데이터정보과학회지 Vol.27 No.4

The big data analysis has received much attention from the researchers working in various fields because the big data has a great potential in detecting or predicting future events such as epidemic outbreaks and changes in stock prices. Reflecting the current popularity of big data analysis, many authors have proposed methods tracking influenza epidemics based on internet-based information. The recently proposed 'autoregressive model using Google (ARGO) model' (Yang et al., 2015) is one of those influenza tracking models that harness search queries from Google as well as the reports from the Centers for Disease Control (CDC), and appears to outperform the existing method such as 'Google Flu Trends (GFT)'. Although the ARGO predicts well the outbreaks of influenza, this study demonstrates that a classical seasonal autoregressive integrated moving average (SARIMA) model can outperform the ARGO. The SARIMA model incorporates more accurate seasonality of the past influenza activities and takes less input variables into account. Our findings show that the SARIMA model is a functional tool for monitoring influenza epidemics.

Comprehensive comparison of normality tests: Empirical study using many different types of data

Lee, Chanmi,Park, Suhwi,Jeong, Jaesik The Korean Data and Information Science Society 2016 한국데이터정보과학회지 Vol.27 No.5

We compare many normality tests consisting of different sources of information extracted from the given data: Anderson-Darling test, Kolmogorov-Smirnov test, Cramervon Mises test, Shapiro-Wilk test, Shaprio-Francia test, Lilliefors, Jarque-Bera test, D'Agostino' D, Doornik-Hansen test, Energy test and Martinzez-Iglewicz test. For the purpose of comparison, those tests are applied to the various types of data generated from skewed distribution, unsymmetric distribution, and distribution with different length of support. We then summarize comparison results in terms of two things: type I error control and power. The selection of the best test depends on the shape of the distribution of the data, implying that there is no test which is the most powerful for all distributions.

The research of new algorithm to improve prediction accuracy of recommender system in electronic commercey

Kim, Sun-Ok The Korean Data and Information Science Society 2010 한국데이터정보과학회지 Vol.21 No.1

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.

On the clustering of huge categorical data

Kim, Dae-Hak The Korean Data and Information Science Society 2010 한국데이터정보과학회지 Vol.21 No.6

Basic objective in cluster analysis is to discover natural groupings of items. In general, clustering is conducted based on some similarity (or dissimilarity) matrix or the original input data. Various measures of similarities between objects are developed. In this paper, we consider a clustering of huge categorical real data set which shows the aspects of time-location-activity of Korean people. Some useful similarity measure for the data set, are developed and adopted for the categorical variables. Hierarchical and nonhierarchical clustering method are applied for the considered data set which is huge and consists of many categorical variables.

Design and evaluation of a fuzzy cooperative caching scheme for MANETs

Bae, Ihn-Han The Korean Data and Information Science Society 2010 한국데이터정보과학회지 Vol.21 No.3

Caching of frequently accessed data in multi-hop ad hoc environment is a technique that can improve data access performance and availability. Cooperative caching, which allows sharing and coordination of cached data among several clients, can further en-hance the potential of caching techniques. In this paper, we propose a fuzzy cooperative caching scheme in mobile ad hoc networks. The cache management of the proposed caching scheme not only uses adaptively CacheData or CachePath based on data sim-ilarity and data utility, but also uses the replacement manager based on data pro t. Also, the proposed caching scheme uses a prefetch manager. When the TTL of the cached data expires, the prefetch manager evaluates the popularity index of the data. If the popularity index is larger than a threshold, the data is prefetched. Otherwise, its space is released. The performance of the proposed scheme is evaluated analytically and is compared to that of other cooperative caching schemes.

A Location Context Management Architecture of Mobile Objects for LBS Application

Ahn, Yoon-Ae Korean Data and Information Science Society 2007 한국데이터정보과학회지 Vol.18 No.4

LBS must manage various context data and make the best use of this data for application service in ubiquitous environment. Conventional mobile object data management architecture did not consider process of context data. Therefore a new mobile data management framework is needed to process location context data. In this paper, we design a new context management framework for a location based application service. A suggestion framework is consisted of context collector, context manager, rule base, inference engine, and mobile object context database. It describes a form of rule base and a movement process of inference engine that are based on location based application scenario. It also presents an embodiment instance of interface which suggested framework is applied to location context interference of mobile object.

Designing Summary Tables for Mining Web Log Data

Ahn, Jeong-Yong Korean Data and Information Science Society 2005 한국데이터정보과학회지 Vol.16 No.1

In the Web, the data is generally gathered automatically by Web servers and collected in server or access logs. However, as users access larger and larger amounts of data, query response times to extract information inevitably get slower. A method to resolve this issue is the use of summary tables. In this short note, we design a prototype of summary tables that can efficiently extract information from Web log data. We also present the relative performance of the summary tables against a sampling technique and a method that uses raw data.

Can a securities law improve investor rationality in processing earnings information?

Kwag, Seung Woog The Korean Data and Information Science Society 2014 한국데이터정보과학회지 Vol.25 No.6

In this paper, I propose a general hypothesis that after the enactment of the Sarbanes-Oxley Act (SOA) financial statements convey more accurate and reliable corporate information to investors who in turn reflect such improvements in stock prices and test four practical hypotheses that simultaneously feature the degree of information asymmetry, forecast bias, and investor reaction to biased earnings information. The empirical results unanimously suggest that the post-SOA investors take advantage of the improvement in informational efficiency and accuracy and actively adjust for analyst forecast bias in earnings forecasts. The SOA indeed appears to achieve its primary goal of investor protection.

A note on Box-Cox transformation and application in microarray data

Rahman, Mezbahur,Lee, Nam-Yong The Korean Data and Information Science Society 2011 한국데이터정보과학회지 Vol.22 No.5

The Box-Cox transformation is a well known family of power transformations that brings a set of data into agreement with the normality assumption of the residuals and hence the response variable of a postulated model in regression analysis. Normalization (studentization) of the regressors is a common practice in analyzing microarray data. Here, we implement Box-Cox transformation in normalizing regressors in microarray data. Pridictabilty of the model can be improved using data transformation compared to studentization.