The emergence of social media in Thailand has given millions of users a platform to express and share their opinions about products and services, among other subjects, and so Twitter is considered to be a rich source of information for companies to understand their customers by extracting and analyzing sentiment from Tweets. The main goal was to investigate the possible uses of Twitter information for businesses in Thailand to take advantage of and to solve any associated limitations caused by the semantics of the Thai language. Hence, social m...

In this study, the two upper bounds of the ruin probability for discrete time risk model derived by adding two controlled factors to the classical discrete time risk model: proportional reinsurance and investment are proposed. These upper bounds are derived using an inductive method and rely on a recursive form of the finite time and/or an integral equation of ultimate (infinite time) of ruin probability which is also derived in this study. Both of the upper bounds are formulated by the assumption that the retention level of reinsurance a...

This research aimed to study the relationship between people’s risk-taking traits, risk preference, and financial risk-taking behavior in the situation where the risk warning statements existed. The study was based on an online experiment conducted in Thailand. The results were taken from 640 participants joining and contributing their answers to the tests. In the context of experimental design, each participant was randomly assigned to different groups in order to investigate the effect of risk warning statements on the relationships. The resu...

In this dissertation, a method of feature selection in machine learning, and more particularly supervised learning is presented. Supervised learning is a machine learning task that infers answers from a training data set. In machine learning, training datasets are employed in order to create a model which enables reasonable predictions, while in supervised learning, each training example is a training set consisting of instances and labels, and the learning objective is to be able to predict the label of a new unseen instance with as few ...

In this dissertation, a new test statistic for testing for a block diagonal covariance matrix structure with a multivariate normal population where the number of variables p exceeds the number of observations n is proposed. Whereas classical approaches such as the likelihood ratio test cannot be applied when p n  , the proposed test statistic is based on the ratio of the estimators of 2 tr and 2 trD , where  is the population covariance matrix and D is the population covariance matrix under the null hypothesis. Furtherm...

In this study, a test statistic used to select a multivariate linear regression model based on Mallows’s Cp with the same rationale as the SCp criterion from the system of equations Vichit Lorchirachoonkul and Jirawan Jitthavech (2012: 2386- 2394) proposed. The D statistic, which is the difference between the modified Cp statistics in the reduced model and in the full model, approximates to a standard normal distribution.

There are two test statistics proposed in this study in order to test whether data come from a gamma distribution. Both of the proposed test statistics are developed from a modified Kendall coefficient based on the independence property of a gamma distribution. The first one is asymptotically distributed as standard normal and the limit distribution function of the second one was improved using an Edgeworth expansion and the Jackknife method. They are invariant to scale parameters and perform substantially better than existing tests in terms of...

High-dimensional data are ubiquitous and bring new challenges, not only to statisticians, but also to researchers in many scientific fields. They arise in situations where the dimension ( p) , the number of variables in a unit, is larger than the sample size (n), the number of units; data analysis using classical multivariate methods can no longer be applied.

Resource discovery is the process to find the queried object also referred as resource searching. With the rapid development of mobile hardware and wireless internet access, it became feasible to use mobile devices in P2P network. Deploying unstructured Peer-to-Peer (P2P) applications over Mobile Ad-hoc Network (MANET) has seen a decade effort, but still has many limitations and challenges. Searching in such resultant networks are not efficient and further ineffective mainly due to mobility, peer discovery and connectivity issues. Typical resou...

In this study, a two- class revenue management ( RM) model, which combines two of the most important RM strategies for a passenger airline: overbooking and seat inventory control is proposed. Using this model, it is possible to concurrently find both the optimal booking limit and the optimal overbooking limit. Consequently, on a closed- form expression for an optimal booking/ overbooking limit, sensitivity analysis was analytically assessed by changing model parameters such as revenue, the penalty cost associated with unsatisfied demand, the sh...

Graphical passwords have been studied by a variety of methods. According to the psychological beliefs, human can remember an image better than text. Therefore, clicking on a graphical password is easier than typing a text password. Moreover, graphical password can avoid the limitation of text password. This research introduces a method for improving the security of authentication by using a blurred image with a cue as a graphical password that is an integration of recognition-based graphical password and cued-recall based graphical passwords to...

Dealing with a large amount of available data becomes a major challenge in data mining and machine learning. Feature selection is a significant preprocessing step for selecting the most informative features by removing irrelevant and redundant features, especially for large datasets. These selected features play an important role in information searching and enhancing the performance of machine learning models such as classification and prediction. There have been several strategies proposed in the past few decades. In this dissertation, we pr...

This mix-method study aims at investigating dimensions of awareness on Green ICT and self-directed learning among employees working in the listed companies in The Securities Exchange of Thailand (SET). The influences of self-directed learning on awareness of Green ICT were also examined. The dimensions of awareness on Green ICT were collected through 37-item Green ICT questionnaire, and the results revealed that there were primarily three dimensions of awareness on Green ICT among Thai employees: 1) Saving energy and Avoiding using wasteful pap...

Nowadays, various industries widely utilize Knowledge Management Systems (KMSs) to facilitate knowledge management activities for improving organizational performance.  Nevertheless, organizations that use KMSs, are still facing difficulties in managing knowledge.  This research intends to enhance the functionalities of KMSs to better facilitate all KM processes.  First, the research aims to identify current problems occurred in the organization during knowledge discovery, knowledge capture, knowledge sharing, and knowledge application processe...

In this dissertation, two studies that are beneficial for actuaries and the insurance business are proposed. In the first study, an exponentiated Weibull distribution using gamma-generated distribution is modified to obtain an alternative gamma-exponentiated Weibull (AGEW) distribution; its sub-models include both gamma and Weibull distributions, both of which are popular in claims modeling by insurance companies. Its basic structural properties such as distribution function, density function, and moments were investigated. Moreover, the ...

In this study, the knowledge of parameter estimation theory based on the corrected score (CS) approach is extended in a linear multivariate multiple regression model with heteroscedastic measurement errors (HME) and an unknown HME variance. The heteroscedasticity of the HME variance is assumed to be capable of being grouped into similar patterns where the sample of observations are assembled into several sub-samples with the property that the variances of the measurement error (ME) are homoscedastic within a group but heteroscedastic betw...

Agreement on word-object pairing in communication depends on the intensity of belief that gradually emerges in a society of agents under the condition that no one is born with embedded knowledge. In know-nothing word-object pairing, the agents in communication find meaning until they reach a consensus on what an object should be called. A language game is a social process of finding agreement on word-object pairing which enables its communication in a multi-agent system. In this research, techniques are proposed to discover the associatio...

Response surface methodology (RSM) is techniques combine both of experimental designs and statistical techniques for empirical model building and optimization. The experimental design is considered by the objective is to optimize one or more response variables influenced by several independent variables. However, in real situation, we may not be able to identify the true model and so an approximated model, usually a central composite design for building a second-order polynomial model, this design is popular in RSM.

A keystroke authentication method has a relatively low cost, yet it is more powerful and easier to use than other biometric authentication methods. However, traditional keystroke authentication has many weaknesses and is easily exploited by malicious actors. Malicious attacks can include shoulder surfing attacks, eavesdropping attacks and key-logger attacks. When users try to access their computer or portable device by using a keystroke authentication method, the users must push the correct buttons with the correct rhythm in order to be authent...

The research aims to identify Composite Leading Indexes (CLIs) and develops the Early Warning System (EWS) by Partial Least Squares Structural Equation Modeling (PLS-SEM). The objective of EWS is to forecast the Economic Cycle (EC) in short-term, medium-term, and long-term periods. Data from Thailand during Q1/2003-Q4/2008 are applied to pursue the purposes. The research uses Real Gross Domestic Product (GDP) as a proxy for the Thai economy, which is the target variable that EWS aims to early signal. The indicators from various economic sect...