• Journal of Intelligence and Information Systems
• /
• v.18 no.4
• /
• pp.43-57
• /
• 2012

To enhance the competitive advantage in a constantly changing business environment, an enterprise management must make the right decision in many business activities based on both internal and external information. Thus, providing accurate information plays a prominent role in management's decision making. Intuitively, historical data can provide a feasible estimate through the forecasting models. Therefore, if the service department can estimate the service quantity for the next period, the service department can then effectively control the inventory of service related resources such as human, parts, and other facilities. In addition, the production department can make load map for improving its product quality. Therefore, obtaining an accurate service forecast most likely appears to be critical to manufacturing companies. Numerous investigations addressing this problem have generally employed statistical methods, such as regression or autoregressive and moving average simulation. However, these methods are only efficient for data with are seasonal or cyclical. If the data are influenced by the special characteristics of product, they are not feasible. In our research, we propose a forecasting framework that predicts service demand of manufacturing organization by combining Case-based reasoning (CBR) and leveraging an unsupervised artificial neural network based clustering analysis (i.e., Self-Organizing Maps; SOM). We believe that this is one of the first attempts at applying unsupervised artificial neural network-based machine-learning techniques in the service forecasting domain. Our proposed approach has several appealing features : (1) We applied CBR and SOM in a new forecasting domain such as service demand forecasting. (2) We proposed our combined approach between CBR and SOM in order to overcome limitations of traditional statistical forecasting methods and We have developed a service forecasting tool based on the proposed approach using an unsupervised artificial neural network and Case-based reasoning. In this research, we conducted an empirical study on a real digital TV manufacturer (i.e., Company A). In addition, we have empirically evaluated the proposed approach and tool using real sales and service related data from digital TV manufacturer. In our empirical experiments, we intend to explore the performance of our proposed service forecasting framework when compared to the performances predicted by other two service forecasting methods; one is traditional CBR based forecasting model and the other is the existing service forecasting model used by Company A. We ran each service forecasting 144 times; each time, input data were randomly sampled for each service forecasting framework. To evaluate accuracy of forecasting results, we used Mean Absolute Percentage Error (MAPE) as primary performance measure in our experiments. We conducted one-way ANOVA test with the 144 measurements of MAPE for three different service forecasting approaches. For example, the F-ratio of MAPE for three different service forecasting approaches is 67.25 and the p-value is 0.000. This means that the difference between the MAPE of the three different service forecasting approaches is significant at the level of 0.000. Since there is a significant difference among the different service forecasting approaches, we conducted Tukey's HSD post hoc test to determine exactly which means of MAPE are significantly different from which other ones. In terms of MAPE, Tukey's HSD post hoc test grouped the three different service forecasting approaches into three different subsets in the following order: our proposed approach > traditional CBR-based service forecasting approach > the existing forecasting approach used by Company A. Consequently, our empirical experiments show that our proposed approach outperformed the traditional CBR based forecasting model and the existing service forecasting model used by Company A. The rest of this paper is organized as follows. Section 2 provides some research background information such as summary of CBR and SOM. Section 3 presents a hybrid service forecasting framework based on Case-based Reasoning and Self-Organizing Maps, while the empirical evaluation results are summarized in Section 4. Conclusion and future research directions are finally discussed in Section 5.

• Journal of Radiation Protection and Research
• /
• v.26 no.2
• /
• pp.59-66
• /
• 2001

It is a hot issue to determine the spatial location and shape of tumor boundary in fractionated stereotactic radiotherapy (FSRT). We could get consecutive transaxial plane images from the phantom (paraffin) and 4 patients with brain tumor using helical computed tomography(HCT). K-means classification algorithm was adjusted to change raw data pixel value in CT images into classified average pixel value. The classified images consists of 5 regions that ate tumor region (TR), normal region (NR), combination region (CR), uncommitted region (UR) and artifact region (AR). The major concern was how to separate the normal region from tumor region in the combination area. Relative average deviation analysis was adjusted to alter average pixel values of 5 regions into 2 regions of normal and tumor region to define maximum point among average deviation pixel values. And then we drawn gross tumor volume (GTV) boundary by connecting maximum points in images using semi-automatic contour method by IDL(Interactive Data Language) program. The error limit of the ROI boundary in homogeneous phantom is estimated within ${\pm}1%$. In case of 4 patients, we could confirm that the tumor lesions described by physician and the lesions described automatically by the K-mean classification algorithm and relative average deviation analyses were similar. These methods can make uncertain boundary between normal and tumor region into clear boundary. Therefore it will be useful in the CT images-based treatment planning especially to use above procedure apply prescribed method when CT images intermittently fail to visualize tumor volume comparing to MRI images.

• Journal of Intelligence and Information Systems
• /
• v.18 no.3
• /
• pp.137-152
• /
• 2012

In the generation of Web 2.0, as many users start to make lots of web contents called user created contents by themselves, the World Wide Web is overflowing by countless information. Therefore, it becomes the key to find out meaningful information among lots of resources. Nowadays, the information retrieval is the most important thing throughout the whole field and several types of search services are developed and widely used in various fields to retrieve information that user really wants. Especially, the legal information search is one of the indispensable services in order to provide people with their convenience through searching the law necessary to their present situation as a channel getting knowledge about it. The Office of Legislation in Korea provides the Korean Law Information portal service to search the law information such as legislation, administrative rule, and judicial precedent from 2009, so people can conveniently find information related to the law. However, this service has limitation because the recent technology for search engine basically returns documents depending on whether the query is included in it or not as a search result. Therefore, it is really difficult to retrieve information related the law for general users who are not familiar with legal terms in the search engine using simple matching of keywords in spite of those kinds of efforts of the Office of Legislation in Korea, because there is a huge divergence between everyday words and legal terms which are especially from Chinese words. Generally, people try to access the law information using everyday words, so they have a difficulty to get the result that they exactly want. In this paper, we propose a term mapping methodology between everyday words and legal terms for general users who don't have sufficient background about legal terms, and we develop a search service that can provide the search results of law information from everyday words. This will be able to search the law information accurately without the knowledge of legal terminology. In other words, our research goal is to make a law information search system that general users are able to retrieval the law information with everyday words. First, this paper takes advantage of tags of internet blogs using the concept for collective intelligence to find out the term mapping relationship between everyday words and legal terms. In order to achieve our goal, we collect tags related to an everyday word from web blog posts. Generally, people add a non-hierarchical keyword or term like a synonym, especially called tag, in order to describe, classify, and manage their posts when they make any post in the internet blog. Second, the collected tags are clustered through the cluster analysis method, K-means. Then, we find a mapping relationship between an everyday word and a legal term using our estimation measure to select the fittest one that can match with an everyday word. Selected legal terms are given the definite relationship, and the relations between everyday words and legal terms are described using SKOS that is an ontology to describe the knowledge related to thesauri, classification schemes, taxonomies, and subject-heading. Thus, based on proposed mapping and searching methodologies, our legal information search system finds out a legal term mapped with user query and retrieves law information using a matched legal term, if users try to retrieve law information using an everyday word. Therefore, from our research, users can get exact results even if they do not have the knowledge related to legal terms. As a result of our research, we expect that general users who don't have professional legal background can conveniently and efficiently retrieve the legal information using everyday words.

• Journal of Intelligence and Information Systems
• /
• v.27 no.3
• /
• pp.139-156
• /
• 2021

The biggest reason for using a deep learning model in image classification is that it is possible to consider the relationship between each region by extracting each region's features from the overall information of the image. However, the CNN model may not be suitable for emotional image data without the image's regional features. To solve the difficulty of classifying emotion images, many researchers each year propose a CNN-based architecture suitable for emotion images. Studies on the relationship between color and human emotion were also conducted, and results were derived that different emotions are induced according to color. In studies using deep learning, there have been studies that apply color information to image subtraction classification. The case where the image's color information is additionally used than the case where the classification model is trained with only the image improves the accuracy of classifying image emotions. This study proposes two ways to increase the accuracy by incorporating the result value after the model classifies an image's emotion. Both methods improve accuracy by modifying the result value based on statistics using the color of the picture. When performing the test by finding the two-color combinations most distributed for all training data, the two-color combinations most distributed for each test data image were found. The result values were corrected according to the color combination distribution. This method weights the result value obtained after the model classifies an image's emotion by creating an expression based on the log function and the exponential function. Emotion6, classified into six emotions, and Artphoto classified into eight categories were used for the image data. Densenet169, Mnasnet, Resnet101, Resnet152, and Vgg19 architectures were used for the CNN model, and the performance evaluation was compared before and after applying the two-stage learning to the CNN model. Inspired by color psychology, which deals with the relationship between colors and emotions, when creating a model that classifies an image's sentiment, we studied how to improve accuracy by modifying the result values based on color. Sixteen colors were used: red, orange, yellow, green, blue, indigo, purple, turquoise, pink, magenta, brown, gray, silver, gold, white, and black. It has meaning. Using Scikit-learn's Clustering, the seven colors that are primarily distributed in the image are checked. Then, the RGB coordinate values of the colors from the image are compared with the RGB coordinate values of the 16 colors presented in the above data. That is, it was converted to the closest color. Suppose three or more color combinations are selected. In that case, too many color combinations occur, resulting in a problem in which the distribution is scattered, so a situation fewer influences the result value. Therefore, to solve this problem, two-color combinations were found and weighted to the model. Before training, the most distributed color combinations were found for all training data images. The distribution of color combinations for each class was stored in a Python dictionary format to be used during testing. During the test, the two-color combinations that are most distributed for each test data image are found. After that, we checked how the color combinations were distributed in the training data and corrected the result. We devised several equations to weight the result value from the model based on the extracted color as described above. The data set was randomly divided by 80:20, and the model was verified using 20% of the data as a test set. After splitting the remaining 80% of the data into five divisions to perform 5-fold cross-validation, the model was trained five times using different verification datasets. Finally, the performance was checked using the test dataset that was previously separated. Adam was used as the activation function, and the learning rate was set to 0.01. The training was performed as much as 20 epochs, and if the validation loss value did not decrease during five epochs of learning, the experiment was stopped. Early tapping was set to load the model with the best validation loss value. The classification accuracy was better when the extracted information using color properties was used together than the case using only the CNN architecture.

• Journal of Global Scholars of Marketing Science
• /
• v.19 no.2
• /
• pp.53-67
• /
• 2009

For the last decades, service quality has been studied as one of the most important tools for a service company to compete with the other companies. Based on these past researches, it has been agreed that the service quality is a basic and powerful tool to create the competitive advantage. Due to similar reason, many service marketing practitioners have been also focused on the service quality to retain the existing consumers and collect the new consumers. However, service quality is subjectively perceived by individual consumers. Consumer evaluation of service quality can be different from each other. Especially consumers with one life-style may evaluate the service quality differently from the consumers with the other life-styles. Therefore we need to know whether there are differences in service quality perception on the categories of life-style. Life-style refers to a distinctive mode of living in its aggregate and broadest sense. It embodies the patterns that were developed and emerged from the dynamics of living in a society. Since the concept of life-style and its relationship to marketing was introduced in 1963 by William Lazer, methods of measuring the life-style and their application have been developed. Life-style has been usually used to segment the marketplace because it offers marketers a unique and important view of the market. When Life-style is combined with clustering methods, life-style segmentation can generate identifiable whole persons rather than isolated fragment. Life-style segmentation begins with people instead of products and classifies them into different life-style types, each characterized by a unique style of living based on a wide range of activities, interests, and opinions(Plummer, 1974). In this study we applies the life-style segmentation based on the AIO(Activities, Interests, and Opinions) to the consumers of the large discount stores. In Korea, the large discount store market has entered into maturity stage so that the market differentiation strategy is becoming a more critical issue to the marketing practitioners. One of the most important tools to differentiate from the competitors in large discount store market is continuously to provide service of better quality than competitors. This study tries to find answers about the following questions: 1) How can we categorize the consumer life-styles in the large discount store? 2) What are the characteristics of the categorized groups? 3) Are there any differences in service quality perception among the consumers with different life-styles 4) Are there any differences in consumer behavior among them in the large discount store? For the purpose, we collected survey data from consumers and analyzed the data with the SPSS package where we had $X^2$-test, factor analysis, ANOVA, MANOVA, and cluster analysis. The survey was made during one month in the April of 2008. Among the collected 306 copies of questionnaires, 281 copies were chosen as the effective samples for empirical analysis except 25 copies with wrong responses. To identify the life-style patterns, we used the measures employed by Kim and Kwon(1999), where 44 items on a seven-point scale were used to measure factors of the life-style patterns. The Principal Component Method was used for factor extraction, and the VARIMAX orthogonal factor rotation was employed. The 7 items showing low factor loading were eliminated. The results of the factor analysis suggested that nine factors of the life-style patterns were identified as follows: 1) the equality-of-sexes and pursuit-of-independence tendency 2) self-management tendency 3) sociable tendency 4) self-display tendency 5) degree of a dilettante life 6) pursuit-of-information tendency 7) bargain hunter tendency 8) TV preference tendency 9) pursuit-of-leisure tendency. Next, after the K-means cluster analysis was performed with nine factors of the life-style patterns, the life-styles of the respondents were classified into four groups which are named as the 'progressive practicality-oriented group', 'positive success-oriented group', 'sociable ostentation-oriented group', 'stable conservation-oriented group'. The analysis results for usage behavior between the market segments showed statistically significant differences in the frequency of usage, duration time in the store, consumer satisfaction, and loyalty. Also, we tried to investigate whether the large discount store consumers differently perceive the quality of service based upon the types of life-style. To measure the service quality of large discount store, we adapted several measurement models measuring the service quality such as SERVPERF, BCP, R-SERVPERF, R-BCP. MANOVA and One-Way ANOVA were performed to confirm the difference in service quality perception based on the market segments. The results have also shown significant differences between life-style types in service quality perception. These findings show that the large discount store marketers should consider consumer life-style as one of the most important market segments for marketing and understand the difference in service quality perception between life-style types. Our findings give important implications to marketers of large discount stores as well as life-style researchers. First, this study showed there were significant differences in consumer's service quality perception and usage behavior between the types of life-style. It provides evidence that the life-style approach can be a important basis in segmenting the large discount store market and will make consumers perceive the service quality high. Second, most previous researches on service quality have been in aggregate level. However, our results imply that the future research on service quality have to focus on segment level.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.45 no.3
• /
• pp.66-79
• /
• 2017

In 2014, a Donghak Peasant Revolution Memorial Park design competition was held with various forms and techniques to convey mourning. This is a process of the reconsideration of memorial projects that are used to stimulate the collective memory and it is a meaningful resource for examining the consciousness of contemporary designers in regards to the memorial designs that are currently under planning in Korea. This study investigated the background of the Donghak commemorative projects that took place at the same site in a timely manner and analyzed the design competition through the existing literature research. Through this, it was seen that the memorial, which was formed by means of past political purposes, has changed into a way to collect various opinions and forms through open design competition. A framework of analysis prepared through multi-layer analysis is daily use, interaction and spontaneity, abstraction, temporality, locality, integration and harmony with surroundings. The results of this study are as follows. First, in order to convey memorial commemoration in everyday life, the projects organized scattered memorial spaces with special characteristics and linked them with daily activities program. Second, the projects used direct participation and emotional experiences to interact with monuments. Third, color, vertical elements, clustering, and park frame manipulation were used for abstract reproduction. Fourth, the projects introduce architecture and furniture that can be changed and plants for temporal change. Fifth, the previous terrain was restored and the setting of the scene was reproduced in order to make the site a space with place. Sixth, to improve the connection with existing monuments, the projects used techniques such as relaxation and the reinforcement of circulation lines and axes. Seventh, a path and a building conforming to the terrain were arranged for harmony with the surroundings.

• Management & Information Systems Review
• /
• v.39 no.2
• /
• pp.181-193
• /
• 2020

This study is a study that analyzes consumer perception by utilizing text mining, which is a recent issue. we analyzed the consumer's perception of Samsung Galaxy by analyzing consumer reviews of Samsung Galaxy YouTube ads. for analysis, 1,819 consumer reviews of YouTube ads were extracted. through this data pre-processing, keywords for advertisements were classified and extracted into nouns, adjectives, and adverbs. after that, frequency analysis and emotional analysis were performed. Finally, clustering was performed through CONCOR. the summary of this study is as follows. the first most frequently mentioned words were Galaxy Note (n = 217), Good (n = 135), Pen (n = 40), and Function (n = 29). it can be judged through the advertisement that consumers "Galaxy Note", "Good", "Pen", and "Features" have good functional aspects for Samsung mobile phone products and positively recognize the Note Pen. in addition, the recognition of "Samsung Pay", "Innovation", "Design", and "iPhone" shows that Samsung's mobile phone is highly regarded for its innovative design and functional aspects of Samsung Pay. second, it is the result of sentiment analysis on YouTube advertising. As a result of emotional analysis, the ratio of emotional intensity was positive (75.95%) and higher than negative (24.05%). this means that consumers are positively aware of Samsung Galaxy mobile phones. As a result of the emotional keyword analysis, positive keywords were "good", "good", "innovative", "highest", "fast", "pretty", etc., negative keywords were "frightening", "I want to cry", "discomfort", "sorry", "no", etc. were extracted. the implication of this study is that most of the studies by quantitative analysis methods were considered when looking at the consumer perception study of existing advertisements. In this study, we deviated from quantitative research methods for advertising and attempted to analyze consumer perception through qualitative research. this is expected to have a great influence on future research, and I am sure that it will be a starting point for consumer awareness research through qualitative research.

• Korean Journal of Poultry Science
• /
• v.50 no.1
• /
• pp.41-50
• /
• 2023

Over the last decade, avian influenza (AI) has been considered an emerging disease that would become the next pandemic, particularly in countries like South Korea, with continuous animal outbreaks. In this situation, risk assessment is highly needed to prevent and prepare for human infection with AI. Thus, we developed the risk assessment matrix for a high-risk area of human infection with AI in South Korea based on the notion that risk is the multiplication of hazards with vulnerability. This matrix consisted of highly pathogenic avian influenza (HPAI) in poultry farms and the number of poultry-associated production facilities assumed as hazards of avian influenza and vulnerability, respectively. The average number of HPAI in poultry farms at the 229-municipal level as the hazard axis of the matrix was predicted using a negative binomial regression with nationwide outbreaks data from 2003 to 2018. The two components of the matrix were classified into five groups using the K-means clustering algorithm and multiplied, consequently producing the area-specific risk level of human infection. As a result, Naju-si, Jeongeup-si, and Namwon-si were categorized as high-risk areas for human infection with AI. These findings would contribute to designing the policies for human infection to minimize socio-economic damages.

• Asia pacific journal of information systems
• /
• v.21 no.1
• /
• pp.103-122
• /
• 2011

Recently, numerous documents have been made available electronically. Internet search engines and digital libraries commonly return query results containing hundreds or even thousands of documents. In this situation, it is virtually impossible for users to examine complete documents to determine whether they might be useful for them. For this reason, some on-line documents are accompanied by a list of keywords specified by the authors in an effort to guide the users by facilitating the filtering process. In this way, a set of keywords is often considered a condensed version of the whole document and therefore plays an important role for document retrieval, Web page retrieval, document clustering, summarization, text mining, and so on. Since many academic journals ask the authors to provide a list of five or six keywords on the first page of an article, keywords are most familiar in the context of journal articles. However, many other types of documents could not benefit from the use of keywords, including Web pages, email messages, news reports, magazine articles, and business papers. Although the potential benefit is large, the implementation itself is the obstacle; manually assigning keywords to all documents is a daunting task, or even impractical in that it is extremely tedious and time-consuming requiring a certain level of domain knowledge. Therefore, it is highly desirable to automate the keyword generation process. There are mainly two approaches to achieving this aim: keyword assignment approach and keyword extraction approach. Both approaches use machine learning methods and require, for training purposes, a set of documents with keywords already attached. In the former approach, there is a given set of vocabulary, and the aim is to match them to the texts. In other words, the keywords assignment approach seeks to select the words from a controlled vocabulary that best describes a document. Although this approach is domain dependent and is not easy to transfer and expand, it can generate implicit keywords that do not appear in a document. On the other hand, in the latter approach, the aim is to extract keywords with respect to their relevance in the text without prior vocabulary. In this approach, automatic keyword generation is treated as a classification task, and keywords are commonly extracted based on supervised learning techniques. Thus, keyword extraction algorithms classify candidate keywords in a document into positive or negative examples. Several systems such as Extractor and Kea were developed using keyword extraction approach. Most indicative words in a document are selected as keywords for that document and as a result, keywords extraction is limited to terms that appear in the document. Therefore, keywords extraction cannot generate implicit keywords that are not included in a document. According to the experiment results of Turney, about 64% to 90% of keywords assigned by the authors can be found in the full text of an article. Inversely, it also means that 10% to 36% of the keywords assigned by the authors do not appear in the article, which cannot be generated through keyword extraction algorithms. Our preliminary experiment result also shows that 37% of keywords assigned by the authors are not included in the full text. This is the reason why we have decided to adopt the keyword assignment approach. In this paper, we propose a new approach for automatic keyword assignment namely IVSM(Inverse Vector Space Model). The model is based on a vector space model. which is a conventional information retrieval model that represents documents and queries by vectors in a multidimensional space. IVSM generates an appropriate keyword set for a specific document by measuring the distance between the document and the keyword sets. The keyword assignment process of IVSM is as follows: (1) calculating the vector length of each keyword set based on each keyword weight; (2) preprocessing and parsing a target document that does not have keywords; (3) calculating the vector length of the target document based on the term frequency; (4) measuring the cosine similarity between each keyword set and the target document; and (5) generating keywords that have high similarity scores. Two keyword generation systems were implemented applying IVSM: IVSM system for Web-based community service and stand-alone IVSM system. Firstly, the IVSM system is implemented in a community service for sharing knowledge and opinions on current trends such as fashion, movies, social problems, and health information. The stand-alone IVSM system is dedicated to generating keywords for academic papers, and, indeed, it has been tested through a number of academic papers including those published by the Korean Association of Shipping and Logistics, the Korea Research Academy of Distribution Information, the Korea Logistics Society, the Korea Logistics Research Association, and the Korea Port Economic Association. We measured the performance of IVSM by the number of matches between the IVSM-generated keywords and the author-assigned keywords. According to our experiment, the precisions of IVSM applied to Web-based community service and academic journals were 0.75 and 0.71, respectively. The performance of both systems is much better than that of baseline systems that generate keywords based on simple probability. Also, IVSM shows comparable performance to Extractor that is a representative system of keyword extraction approach developed by Turney. As electronic documents increase, we expect that IVSM proposed in this paper can be applied to many electronic documents in Web-based community and digital library.

• Journal of Global Scholars of Marketing Science
• /
• v.17 no.2
• /
• pp.1-29
• /
• 2007

The flow theory becomes one of the most important frameworks in the internet research arena. Hoffman and Novak proposed a hierarchical flow model showing the antecedents and outcomes of flow and the relationship among these variables in the hyper-media computer circumstances (Hoffman and Novak 1996). This model was further tested after their initial research (Novak, Hoffman, and Yung 2000). At their paper, Hoffman and Novak explained that the balance of challenge and skill leads to flow which means the positive optimal state of mind (Hoffman and Novak 1996). An imbalance between challenge and skill, leads to negative states of mind like anxiety, boredom, apathy (Csikszentmihalyi and Csikszentmihalyi 1988). Almost all research on the flow 4-channel model have been focusingon flow, the positive state of mind (Ellis, Voelkl, and Morris 1994 Mathwick and Rigdon 2004). However, it also needs to examine the formation of the negative states of minds and their outcomes. Flow researchers explain play or playfulness as antecedents or the early state of flow. However, play has been regarded as a distinct concept from flow in the flow literatures (Hoffman and Novak 1996; Novak, Hoffman, and Yung 2000). Mathwick and Rigdon discovered the influences of challenge and skill on play; they also observed the influence of play on web-loyalty and brand loyalty (Mathwick and Rigdon 2004). Unfortunately, they did not go so far as to test the influences of play on state of mind. This study focuses on the relationships between state of mind in the flow 4-channel model and play. Early research has attempted to hypothetically explain state of mind in flow theory, but has not been tested except flow until now. Also the importance of play has been emphasized in the flow theory, but has not been tested in the flow 4-channel model context. This researcher attempts to analyze the relationships among state of mind, skill of play, challenge, state of mind and web loyalty. For this objective, I developed a measure for state of mind and defined the concept of play as a trait. Then, the influences of challenge and skill on the state of mind and play under on-line shopping conditions were tested. Also the influences of play on state of mind were tested and those of flow and play on web loyalty were highlighted. 294 undergraduate students participated in this research survey. They were asked to respond about their perceptions of challenge, skill, state of mind, play, and web-loyalty to on-line shopping mall. Respondents were restricted to students who bought products on-line in a month. In case of buying products at two or more on-line shopping malls, they asked to respond about the shopping mall where they bought the most important one. Construct validity, discriminant validity, and convergent validity were used to check the measurement validations. Also, Cronbach's alpha was used to check scale reliability. A series of exploratory factor analyses was conducted. This researcher conducted confirmatory factor analyses to assess the validity of measurements. All items loaded significantly on their respective constructs. Also, all reliabilities were greater than.70. Chi-square difference tests and goodness of fit tests supported discriminant and convergent validity. The results of clustering and ANOVA showed that high challenge and high skill leaded to flow, low challenge and high skill leaded to boredom, and low challenge and low skill leaded to apathy. But, it was different from my expectation that high challenge and low skill didnot lead to anxiety but leaded to apathy. The results also showed that high challenge and high skill, and high challenge and low skill leaded to the highest play. Low challenge leaded to low play. 4 Structural Equation Models were built by flow, anxiety, boredom, apathy for analyzing not only the impact of play on state of mind and web-loyalty, but also that of state of mind on web-loyalty. According the analyses results of these models, play impacted flow and web-loyalty positively, but impacted anxiety, boredom, and apathy negatively. Results also showed that flow impacted web-loyalty positively, but anxiety, boredom, and apathy impacted web-loyalty negatively. The interpretations and implications of the test results of the hypotheses are as follows. First, respondents belonging to different clusters based on challenge and skill level experienced different states of mind such as flow, anxiety, boredom, apathy. The low challenge and low skill group felt the highest anxiety and apathy. It could be interpreted that this group feeling high anxiety or fear, then avoided attempts to shop on-line. Second, it was found that higher challenge leads to higher levels of play. Test results show that the play level of the high challenge and low skill group (anxiety group) was higher than that of the high challenge and high skill group (flow group). However, this was not significant. Third, play positively impacted flow and negatively impacted boredom. The negative impacts on anxiety and apathy were not significant. This means that the combination of challenge and skill creates different results. Forth, play and flow positively impacted web-loyalty, but anxiety, boredom, apathy had negative impacts. The effect of play on web-loyalty was stronger in case of anxiety, boredom, apathy group than fl ow group. These results show that challenge and skill influences state of mind and play. Results also demonstrate how play and flow influence web-loyalty. It implies that state of mind and play should be the core marketing variables in internet marketing. The flow theory has been focusing on flow and on the positive outcomes of flow experiences. But, this research shows that lots of consumers experience the negative state of mind rather than flow state in the internet shopping circumstance. Results show that the negative state of mind leads to low or negative web-loyalty. Play can have an important role with the web-loyalty when consumers have the negative state of mind. Results of structural equation model analyses show that play influences web-loyalty positively, even though consumers may be in the negative state of mind. This research found the impacts of challenge and skill on state of mind in the flow 4-channel model, not only flow but also anxiety, boredom, apathy. Also, it highlighted the role of play in the flow 4-channel model context and impacts on web-loyalty. However, tests show a few different results from hypothetical expectations such as the highest anxiety level of apathy group and insignificant impacts of play on anxiety and apathy. Further research needs to replicate this research and/or to compare 3-channel model with 4-channel model.