2012 13th ACIS International Conference on Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing, 2012
ABSTRACT Text data are a major type of data in the modern society. Literatures have pointed out t... more ABSTRACT Text data are a major type of data in the modern society. Literatures have pointed out that more than 80% of data are in text form. It is important to study the insights from text data in addition to quantitative data. The development of text mining techniques started in the early 80's. The methodology has become much more mature in the recent years. In this article, we conduct a case study using text mining technique to analyze the patterns of the president's State of the Union address in USA. The speeches analyzed include the recent four USA presidents, Bush (1989-1992), Clinton (1993 - 2000), G.W. Bush (2001-2008), and Obama (2009-2011). Using two different text mining techniques, we identify six clusters from the 23 speeches using one technique and obtain seven topics based on the other technique.
Qualitative research methodology is commonly used for education research. The main advantage is t... more Qualitative research methodology is commonly used for education research. The main advantage is that it is capable of gaining more rigorous insights of research objectives when compared with quantitative research. However, due to the extensive time required, the sample size for a qualitative research is usually very small, and can be as small as two subjects. As a consequence, the results are difficult to be inferred to the population. Text mining methodology is a modern technique to analyze text data. It has been successfully applied to a wide variety of business problems. However, it has not yet been applied as a tool for qualitative research in education. In this study, we attempt to apply text mining method to analyze students' written reasons of a question related to the concept of variation. Four sections of an introductory statistics course participated in the study (n = 218). The results suggest text mining method can be an alternative, if applied appropriately. The stre...
Etezadi-Amoli and Ciampi (Biometrics 1987) introduced a method to approximate the baseline hazard... more Etezadi-Amoli and Ciampi (Biometrics 1987) introduced a method to approximate the baseline hazard and to estimate the regression coefficients for covariates simultaneously for the extended hazard regression (EHR) model using a quadratic spline approach. In this paper, an estimate of the baseline hazard function by B-spline approximation using a minimax criterion is proposed. The nonlinear problem is approximated by a linear programming problem with only linear constraints. The nice features of this approach are: (i) the minimax criterion provides a robust approximation to the hazard function, and (ii) the linearized problem is numerically simple and fast.
... Ghada Sharaf El Din, Carl Lee, and Moataz Fattah1 Central Michigan University, Mt. ... X PY i... more ... Ghada Sharaf El Din, Carl Lee, and Moataz Fattah1 Central Michigan University, Mt. ... X PY i X X β β β β + = = + + where i = 0,1,2 ,10. The weights for the input variables are the same for each level of Y, and the bias terms 0i β 's are different. ... of hidden layers .i H defined as ( ) . . ...
194 Information Technology in Educational Research and Statistics The use of computational techno... more 194 Information Technology in Educational Research and Statistics The use of computational technology in college introductory statis-tics is now very common. The simplest and most commonly used com-putational technology is the scientific calculator, which has been used for ...
In this paper, some properties of gamma-X family are discussed and a member of the family, the ga... more In this paper, some properties of gamma-X family are discussed and a member of the family, the gamma-normal distribution, is studied in detail. The limiting behaviors, moments, mean deviations, dispersion, and Shannon entropy for the gamma-normal distribution are provided. Bounds for the non-central moments are obtained. The method of maximum likelihood estimation is proposed for estimating the parameters of the gamma-normal distribution. Two real data sets are used to illustrate the applications of the gamma-normal distribution.
Two sufficient conditions for characterizing Pitman nearness (PN) are derived. It is found that o... more Two sufficient conditions for characterizing Pitman nearness (PN) are derived. It is found that one of the conditions is a standardized decision criterion. Two examples are given, including comparison of the James-Stein estimator with the usual unbiased estimator. It is also shown that stochastic domination implies PN for comparing any two multivariate normal estimators. Moreover, they are equivalent if two estimators are unbiased.
Abstract: The assumption that is usually made when modeling count data is that the response varia... more Abstract: The assumption that is usually made when modeling count data is that the response variable, which is the count, is correctly reported. Some counts might be over-or under-reported. We derive the Generalized Poisson-Poisson mixture regression (GPPMR) ...
2012 13th ACIS International Conference on Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing, 2012
ABSTRACT Text data are a major type of data in the modern society. Literatures have pointed out t... more ABSTRACT Text data are a major type of data in the modern society. Literatures have pointed out that more than 80% of data are in text form. It is important to study the insights from text data in addition to quantitative data. The development of text mining techniques started in the early 80's. The methodology has become much more mature in the recent years. In this article, we conduct a case study using text mining technique to analyze the patterns of the president's State of the Union address in USA. The speeches analyzed include the recent four USA presidents, Bush (1989-1992), Clinton (1993 - 2000), G.W. Bush (2001-2008), and Obama (2009-2011). Using two different text mining techniques, we identify six clusters from the 23 speeches using one technique and obtain seven topics based on the other technique.
Qualitative research methodology is commonly used for education research. The main advantage is t... more Qualitative research methodology is commonly used for education research. The main advantage is that it is capable of gaining more rigorous insights of research objectives when compared with quantitative research. However, due to the extensive time required, the sample size for a qualitative research is usually very small, and can be as small as two subjects. As a consequence, the results are difficult to be inferred to the population. Text mining methodology is a modern technique to analyze text data. It has been successfully applied to a wide variety of business problems. However, it has not yet been applied as a tool for qualitative research in education. In this study, we attempt to apply text mining method to analyze students' written reasons of a question related to the concept of variation. Four sections of an introductory statistics course participated in the study (n = 218). The results suggest text mining method can be an alternative, if applied appropriately. The stre...
Etezadi-Amoli and Ciampi (Biometrics 1987) introduced a method to approximate the baseline hazard... more Etezadi-Amoli and Ciampi (Biometrics 1987) introduced a method to approximate the baseline hazard and to estimate the regression coefficients for covariates simultaneously for the extended hazard regression (EHR) model using a quadratic spline approach. In this paper, an estimate of the baseline hazard function by B-spline approximation using a minimax criterion is proposed. The nonlinear problem is approximated by a linear programming problem with only linear constraints. The nice features of this approach are: (i) the minimax criterion provides a robust approximation to the hazard function, and (ii) the linearized problem is numerically simple and fast.
... Ghada Sharaf El Din, Carl Lee, and Moataz Fattah1 Central Michigan University, Mt. ... X PY i... more ... Ghada Sharaf El Din, Carl Lee, and Moataz Fattah1 Central Michigan University, Mt. ... X PY i X X β β β β + = = + + where i = 0,1,2 ,10. The weights for the input variables are the same for each level of Y, and the bias terms 0i β 's are different. ... of hidden layers .i H defined as ( ) . . ...
194 Information Technology in Educational Research and Statistics The use of computational techno... more 194 Information Technology in Educational Research and Statistics The use of computational technology in college introductory statis-tics is now very common. The simplest and most commonly used com-putational technology is the scientific calculator, which has been used for ...
In this paper, some properties of gamma-X family are discussed and a member of the family, the ga... more In this paper, some properties of gamma-X family are discussed and a member of the family, the gamma-normal distribution, is studied in detail. The limiting behaviors, moments, mean deviations, dispersion, and Shannon entropy for the gamma-normal distribution are provided. Bounds for the non-central moments are obtained. The method of maximum likelihood estimation is proposed for estimating the parameters of the gamma-normal distribution. Two real data sets are used to illustrate the applications of the gamma-normal distribution.
Two sufficient conditions for characterizing Pitman nearness (PN) are derived. It is found that o... more Two sufficient conditions for characterizing Pitman nearness (PN) are derived. It is found that one of the conditions is a standardized decision criterion. Two examples are given, including comparison of the James-Stein estimator with the usual unbiased estimator. It is also shown that stochastic domination implies PN for comparing any two multivariate normal estimators. Moreover, they are equivalent if two estimators are unbiased.
Abstract: The assumption that is usually made when modeling count data is that the response varia... more Abstract: The assumption that is usually made when modeling count data is that the response variable, which is the count, is correctly reported. Some counts might be over-or under-reported. We derive the Generalized Poisson-Poisson mixture regression (GPPMR) ...
Uploads
Papers by Carl Lee