• Journal of Distribution Science
• /
• v.9 no.4
• /
• pp.5-13
• /
• 2011

This study provides analysis of the effectiveness of domestic marketing strategies of the Korean coffee shop "Caffe Bene". It bases its evaluation on statistical outputs of 'choice attributes,' "market segmentation," demographic characteristics," and "satisfaction differences." The results are summarized in four points. First, five choice attributes were extracted from factor analysis: price, atmosphere, comfort, taste, and location; these are related to coffee shop selection behavior. Based on these five factors, cluster analysis was conducted, with statistical results classifying customers into three major groups: atmosphere oriented; comfort oriented; and taste oriented. Second, discriminant analysis tested cluster analysis and showed two discriminant functions: location and atmosphere. Third, cross-tabulation analysis based on demographic characteristics showed distinctive demographic characteristics within the three groups. Atmosphere oriented group, early-20s, as women of all ages was found to be 'walking down the street 'and 'through acquaintances' in many cases, as the cognitive path, and mostly found the store through 'outdoor advertising', and 'introduction'. Comfort oriented group was mainly women who are students in their early twenties or professionals, and appeared as a group to be very loyal because of high recommendation to other customers compared to other groups. Taste oriented group, unlike the other group, was mainly late-20s' college graduates, and was confirmed, as low loyalty, with lower recommendation activity. Fourth, to analyze satisfaction differences, one-way ANOVA was conducted. It shows that groups which show high satisfaction in the five main factors also show high menu satisfaction and high overall satisfaction. This results show that segmented marketing strategies are necessary because customers are considering price, atmosphere, comfort, taste, location when they choose coffee shop and demographics show different attributes based on segmented groups. For example, atmosphere oriented group is satisfied with shop interior and comfort while dissatisfied with price because most of the customers in this group are early 20s and do not have great financial capability. Thus, price discounting marketing strategies based on individual situations through CRM system is critical. Comfort oriented group shows high satisfaction level about location and shop comfort. Also, in this group, there are many early 20s female customers, students, and self-employed people. This group customers show high word of mouth tendency, hence providing positive brand image to the customers would be important. In case of taste oriented group, while the scores of taste and location are high, word of mouth score is low. This group is mainly composed of educated and professional many late 20s customers, therefore, menu differentiation, increasing quality of coffee taste and price discrimination is critical to increase customers' satisfaction. However, it is hard to generalize the results of study to other coffee shop brand, because this study have researched only one domestic coffee shop, Caffe Bene. Thus if future study expand the scope of locations, brands, and occupations, the results of the study would provide more generalizable results. Finally, research of customer satisfactions of menu, trust, loyalty, and switching cost would be critical in the future study.

• Journal of Intelligence and Information Systems
• /
• v.24 no.1
• /
• pp.167-181
• /
• 2018

Over the past decade, deep learning has been in spotlight among various machine learning algorithms. In particular, CNN(Convolutional Neural Network), which is known as the effective solution for recognizing and classifying images or voices, has been popularly applied to classification and prediction problems. In this study, we investigate the way to apply CNN in business problem solving. Specifically, this study propose to apply CNN to stock market prediction, one of the most challenging tasks in the machine learning research. As mentioned, CNN has strength in interpreting images. Thus, the model proposed in this study adopts CNN as the binary classifier that predicts stock market direction (upward or downward) by using time series graphs as its inputs. That is, our proposal is to build a machine learning algorithm that mimics an experts called 'technical analysts' who examine the graph of past price movement, and predict future financial price movements. Our proposed model named 'CNN-FG(Convolutional Neural Network using Fluctuation Graph)' consists of five steps. In the first step, it divides the dataset into the intervals of 5 days. And then, it creates time series graphs for the divided dataset in step 2. The size of the image in which the graph is drawn is $40(pixels){\times}40(pixels)$, and the graph of each independent variable was drawn using different colors. In step 3, the model converts the images into the matrices. Each image is converted into the combination of three matrices in order to express the value of the color using R(red), G(green), and B(blue) scale. In the next step, it splits the dataset of the graph images into training and validation datasets. We used 80% of the total dataset as the training dataset, and the remaining 20% as the validation dataset. And then, CNN classifiers are trained using the images of training dataset in the final step. Regarding the parameters of CNN-FG, we adopted two convolution filters ($5{\times}5{\times}6$ and $5{\times}5{\times}9$) in the convolution layer. In the pooling layer, $2{\times}2$ max pooling filter was used. The numbers of the nodes in two hidden layers were set to, respectively, 900 and 32, and the number of the nodes in the output layer was set to 2(one is for the prediction of upward trend, and the other one is for downward trend). Activation functions for the convolution layer and the hidden layer were set to ReLU(Rectified Linear Unit), and one for the output layer set to Softmax function. To validate our model - CNN-FG, we applied it to the prediction of KOSPI200 for 2,026 days in eight years (from 2009 to 2016). To match the proportions of the two groups in the independent variable (i.e. tomorrow's stock market movement), we selected 1,950 samples by applying random sampling. Finally, we built the training dataset using 80% of the total dataset (1,560 samples), and the validation dataset using 20% (390 samples). The dependent variables of the experimental dataset included twelve technical indicators popularly been used in the previous studies. They include Stochastic %K, Stochastic %D, Momentum, ROC(rate of change), LW %R(Larry William's %R), A/D oscillator(accumulation/distribution oscillator), OSCP(price oscillator), CCI(commodity channel index), and so on. To confirm the superiority of CNN-FG, we compared its prediction accuracy with the ones of other classification models. Experimental results showed that CNN-FG outperforms LOGIT(logistic regression), ANN(artificial neural network), and SVM(support vector machine) with the statistical significance. These empirical results imply that converting time series business data into graphs and building CNN-based classification models using these graphs can be effective from the perspective of prediction accuracy. Thus, this paper sheds a light on how to apply deep learning techniques to the domain of business problem solving.

• Journal of Intelligence and Information Systems
• /
• v.24 no.3
• /
• pp.1-19
• /
• 2018

Large amount of data is now available for research and business sectors to extract knowledge from it. This data can be in the form of unstructured data such as audio, text, and image data and can be analyzed by deep learning methodology. Deep learning is now widely used for various estimation, classification, and prediction problems. Especially, fashion business adopts deep learning techniques for apparel recognition, apparel search and retrieval engine, and automatic product recommendation. The core model of these applications is the image classification using Convolutional Neural Networks (CNN). CNN is made up of neurons which learn parameters such as weights while inputs come through and reach outputs. CNN has layer structure which is best suited for image classification as it is comprised of convolutional layer for generating feature maps, pooling layer for reducing the dimensionality of feature maps, and fully-connected layer for classifying the extracted features. However, most of the classification models have been trained using online product image, which is taken under controlled situation such as apparel image itself or professional model wearing apparel. This image may not be an effective way to train the classification model considering the situation when one might want to classify street fashion image or walking image, which is taken in uncontrolled situation and involves people's movement and unexpected pose. Therefore, we propose to train the model with runway apparel image dataset which captures mobility. This will allow the classification model to be trained with far more variable data and enhance the adaptation with diverse query image. To achieve both convergence and generalization of the model, we apply Transfer Learning on our training network. As Transfer Learning in CNN is composed of pre-training and fine-tuning stages, we divide the training step into two. First, we pre-train our architecture with large-scale dataset, ImageNet dataset, which consists of 1.2 million images with 1000 categories including animals, plants, activities, materials, instrumentations, scenes, and foods. We use GoogLeNet for our main architecture as it has achieved great accuracy with efficiency in ImageNet Large Scale Visual Recognition Challenge (ILSVRC). Second, we fine-tune the network with our own runway image dataset. For the runway image dataset, we could not find any previously and publicly made dataset, so we collect the dataset from Google Image Search attaining 2426 images of 32 major fashion brands including Anna Molinari, Balenciaga, Balmain, Brioni, Burberry, Celine, Chanel, Chloe, Christian Dior, Cividini, Dolce and Gabbana, Emilio Pucci, Ermenegildo, Fendi, Giuliana Teso, Gucci, Issey Miyake, Kenzo, Leonard, Louis Vuitton, Marc Jacobs, Marni, Max Mara, Missoni, Moschino, Ralph Lauren, Roberto Cavalli, Sonia Rykiel, Stella McCartney, Valentino, Versace, and Yve Saint Laurent. We perform 10-folded experiments to consider the random generation of training data, and our proposed model has achieved accuracy of 67.2% on final test. Our research suggests several advantages over previous related studies as to our best knowledge, there haven't been any previous studies which trained the network for apparel image classification based on runway image dataset. We suggest the idea of training model with image capturing all the possible postures, which is denoted as mobility, by using our own runway apparel image dataset. Moreover, by applying Transfer Learning and using checkpoint and parameters provided by Tensorflow Slim, we could save time spent on training the classification model as taking 6 minutes per experiment to train the classifier. This model can be used in many business applications where the query image can be runway image, product image, or street fashion image. To be specific, runway query image can be used for mobile application service during fashion week to facilitate brand search, street style query image can be classified during fashion editorial task to classify and label the brand or style, and website query image can be processed by e-commerce multi-complex service providing item information or recommending similar item.

• The Korean Journal of Air & Space Law and Policy
• /
• v.29 no.2
• /
• pp.3-54
• /
• 2014

The Republic of Korea enacted the Air Transport Act in Commercial Law which was entered into force in November, 2011. The Air Transport Act in Korean Commercial Law was established to regulate domestic carriage by air and damages to the third party which occur within the territorial area caused by aircraft operations. There are some problems to be reformed in the Provisions of Korean Commercial Law for the aircraft operator's liability of compensation for damages to the third party caused by aircraft operation as follows. First, the aircraft operator's liability of compensation for damages needs to be improved because it is too low to compensate adequately to the third party damaged owing to the aircraft operation. Therefore, the standard of classifying per aircraft weight is required to be detailed from the current 4-tier into 10-tier and the total limited amount of liability is also in need of being increased to the maximum 7-hundred-million SDR. In addition, the limited amount of liability to the personal damage is necessary to be risen from the present 125,000 SDR to 625,000 SDR according to the recent rate of prices increase. This is the most desirable way to improve the current provisions given the ordinary insurance coverage per one aircraft accident and various specifications of recent aircraft in order to compensate the damaged appropriately. Second, the aircraft operator shall be liable without fault to damages caused by terrorism such as hijacking, attacking an aircraft and utilizing it as means of attack like the 9 11 disaster according to the present Air Transport Act in Korean Commercial Law. Some argue that it is too harsh to aircraft operators and irrational, but given they have also some legal duties of preventing terrorism and in respect of helping the third party damaged, it does not look too harsh or irrational. However, it should be amended into exempting aircraft operator's liability when the terrorism using of an aircraft by well-organized terrorists group happens like 9 11 disaster in view of balancing the interest between the aircraft operator and the third party damaged. Third, considering the large scale of the damage caused by the aircraft operation usually aircraft accident, it is likely that many people damaged can be faced with a financial crisis, and the provision of advance payment for air carrier's liability of compensation also needs to be applied to the case of aircraft operator's liability. Fourth, the aircraft operator now shall be liable to the damages which occur in land or water except air according to the current Air Transport Act of Korean Commercial Law. However, because the damages related to the aircraft operation in air caused by another aircraft operation are not different from those in land or water. Therefore, the term of 'on the surface' should be eliminated in the term of 'third parties on the surface' in order to make the damages by the aircraft operation in air caused by another aircraft operation compensable by Air Transport Act of Korean Commercial Law. It is desired that the Air Transport Act in Commercial Law including the clauses related to the aircraft operator's liability of compensation for damages to the third party be developed continually through the resolutions about its problems mentioned above for compensating the third party damaged appropriately and balancing the interest between the damaged and the aircraft operator.

• Journal of the Korean Society of Costume
• /
• v.1
• /
• pp.119-129
• /
• 1977

It is significant for developing the future for us to know our present age. In order to preserve our Korean costume as a fola clothes retaining our distinguished independent characterisitics and to help design the tomorrow of our Korean costume playing a role as a racial to develop the world clothing culture, a survey was conducted to investigate modern people's conscious-ness and wearing practumes of Korean costume by questionaire and interviewing methods. The results of the survey were analyzed as follows: (1) At present, Korean costumes were purchased as customtailored(64.0%) and as ready-made(17.8%) and most of them were not made at individual homes. The laundry and ironing of them were carried out at laundry shops(68.8%). Considering our present economic, social and cultural aspects, sowing, laundryand ironing will not be carried out at homes again in the future and ready made costumes seen to be produced in a large scale in the future. Garment makers and laundry shop operators should be trained how to make our Korean costumes retain our traditional beauty in the course of their production and laundry and the makers of ready-made costumes must make research how to efficiently produce ideal ready-made costumes by adopting the synchro system in their wrk odisivion. (2) The age group wearing Korean costumes most frequently was the aged people over 60 (their wearing rate; 45%-50%) and the group wearing them most frequently next io the aged people over 60, was housewives(their wearing rate; 15%-20%). Excludign aged people and housewives, other respondentsdid not wear Korean costumes very frequently. Men's wearing rate was lower their wearing rate was the younger their ages were and the less their monthly incomes were. Korean costumes were used for holiday and festival(60%), wedding and funeral ceremonies (52%), visiting and working(22%), casual wear(12.8%) and home wear(9.2%). The use of Korean costumes as casual and home wears, was lower than the use for holday, festival, visiting and working, Under our present circumstances in which our Korean people use both Western style clothes and Korean costumer, our Korean costume has lostits position as a basic and necessary requiement in Korean people's daily life and become a ceremonical and fancy costume. It is natural that the times and life change everything in our daily life. Our costume has to be made as good ceremonial and fancy clothes satisfying modern sensibility according to its new role. In order for us to get close with our clothes, a keen study must be carried out to cleat the color, material, style, function and harmony of the Korean costume matching the of the times. (3) The 47.8% of the respondents answered that they were proud of our Korean costume as our folk clothes, 47.6% replied that thought them just common and 1.1% responded that they were ashamed of it. Most of them were affirmative in feeling pride with our Korean costume. (4) Considering the functional aspect of Korean costumes, their strong points were symetric beauty, rhythmical beauty, unity feeling, harmonical beauty and detailed decorations. Their common shortcomings were lack of individuality and inadequateness for active life. The shortcomings of woman costumes were suppressing breast, making resperation difficult and in adequnteness in summer time. The main reason not to wear our Korean costumes, was due to the fact that they are incomvenient for active life. As a measure to eliminate such shortcomings, 1) the suspension system of skirt to remove the suppression of breast should be generally adopted. 2) they should be simplified in their structure to make them convenient for active life and adepuate in wearing them in hot weather in an extent to which the traditional beauty of the costume may not be lostand 3) a new technique must be explored for showing individuality by wearing method and new arrangment of colors and decorations. (5) The reasons desiring to wear Korean costumes were classifide as follows: A. Korean costumes are our traditional clothes(43.4%). B. Korean costumes are noble and beautiful(26.8%). C. They are accustomed to wear Korean costumes by habit(19.5%). D. Korean costumes are necessary for attending ceremoneis(9.5%). E. Miscellaneous reasons(0.8%). Classifying these reasons into age groups, the high age group over 40 wore them because they were easy to wear by habit and the low age group of 10-30 never thought that they were east to wear by habit. Considering that even those who were accustomed to wear Korean costumes showed a low wearing rate and that the young generation were accustomed to wear Western style clothes rather than Korean costumes, the wearing rate of Korean costumes will be reduced in the future if such trend continues. It is urgent for us to make our best efforts in order to enhance the interest of young generation in Korean costumes and not to make them lose the strong points of Korean costume in the future. (6) Conicering the plan of the respondents on what kind of clothes they were going to wear in the future, among the age group over 50, those who wanted to wear only Korean costumes were 24.8%(men) and 35.1%(women), those who wanted to wear 49.7%(men) and 47.4(women), those who wanted to wear chiefly Western style clothes were 20.7% (men) and 14.4%(women) and those who wanted to wear only Western style clothes, were 2.4% (men) and 2.1%(women). This shows that the general tendency to wear only or chiefly Korean costumes is more prevalent than that to wear only Western style. Among the age group under 50, the tendency to wear Western style clothes was conspicuous and most of the respondent answered that they would wear chiefly Western style clothes and Korean costumes occasionally. Only 5.4% of the respondent answered that they would wear only Western style clothes and this shows that meny respondents still wonted to wear Korean costumes. Those who wanted their descendants to wear what they desire, were 50.1%(men) and 68.8% (women) and those who wanted their descendants to wear Koran costumes occasionally, were 85.8%(men) and 86.3%(women). This shows that most of respondents wanted their descendants to wear Korean costumes. In order to realize, it is necessory for us to make ourdescendants recognize the preciousness of our traditional culture and modify our Korean costumes according to their taste so that they may like wearing them.

• The Korean Journal of Nuclear Medicine Technology
• /
• v.18 no.1
• /
• pp.76-81
• /
• 2014

Purpose: Diversity and the lachrymal duct deformities and the passage inside the nasal cavity except for anterior image such as epiphora happens during the test were able to express more precisely during the dacryoscintigraphy. Also, we thought about the necessity of a method to classify the passage into the naso-lachrymal duct from epiphora. Therefore, we are to find the validity of the method to obtain both oblique views except for anterior views. Materials and Methods: The targets of this research are 78 patients with epiphora due to the blockage at the lachrymal duct from January 2013 to August 2013. Average age was $56.96{\pm}13.36$. By using a micropipette, we dropped 1-2 drops of $^{99m}TcO4^-$ of 3.7 MBq (0.1 mCi) with $10{\mu}L$ of each drop into the inferior conjunctival fold, then we performed dynamic check for 20 minutes with 20 frames of each minute. In case of we checked the passage from both eyes to nasal cavity immediately after the dynamic check, we obtained oblique view immediately. If we didn't see the passage in either side of the orbit, we obtained oblique views of the orbit after checking the frontal film in 40 minutes. The instrument we used was Pin-hole Collimator with Gamma Camera(Siemens Orbiter, Hoffman Estates, IL, USA). Results: Among the 78 patients with dacryoscintigraphy, 35 patients were confirmed with passage into the nasal cavity from the anterior view. Among those 35 patients, 15 patients were confirmed with passage into the nasal cavity on both eyes, and it was able to observe better passage patterns through oblique view with a result of 8 on both eyes, 2 on left eye, and 1 on right eye. 20 patients had passage in left eye or right eye, among those patients 10 patients showed clear passage compared to the anterior view. 13 patients had possible passage, and 30 patients had no proof of motion of the tracer. To sum up, 21 patients (60%) among 35 patients showed clear pattern of passage with additional oblique views compared to anterior view. People responded obtaining oblique views though 5 points scale about the utility of passage identification helps make diagnoses the passage, passage delayed, and blockage of naso-lachrymal duct by showing the well-seen portions from anterior view. Also, when classifying passage to naso-lachrymal duct and flow to the skin, oblique views has higher chance of classification in case of epiphora (anterior:$4.14{\pm}0.3$, oblique:$4.55{\pm}0.4$). Conclusion: It is considered that if you obtain oblique views of the bilateral orbits in addition to anterior view during the dacryoscintigraphy, the ability of diagnose for reading will become higher because you will be able to see the areas that you could not observe from the anterior view so that you can see if it emitted after the naso-lachrymal duct and the flow of epiphora on the skin.

• Journal of the Korean Society of Food Culture
• /
• v.20 no.5
• /
• pp.608-614
• /
• 2005

The object of this research is to analyze and classify the dining-out behavior and preference on Korean food by age groups and to make counter proposals for better marketing and planning strategies. Major dining out motives were lack of time, the easiness of preparation, and schedule. For lunch, the schedule was the major dining-out motive. For dinner, the respondents in their 30s and below answered social gathering was their major dining-out motive (40.7% and 31.3% respectively). On the other hand, for the respondents in their 40s and 50s, the family gathering was the major dining motive (50.4% and 55.3% respectively) (${\chi}^{2}=68.081,\;p<0.001$). For dining out frequency, 1-2 dining out per a week had the highest percentage, among which the respondents in their 30s was 42.9% (the highest) and the respondents in their 50s was 18% (the lowest). For the dining-out cost, the respondents in their 30s and below spent more on dinner rather than breakfast or lunch. For the menu preference of Korean foods, Doenjangjigae had the highest percentage. In case of Kimchi, the respondents in their 40s showed higher preference than the respondents in their 30s. Interestingly, the preference for Kimchi was higher in the respondents younger than 30 rather than in the respondents in their 30s. and the respondents older than 40 (p<0.05). Preference for Jangachi was considerably low in the respondents younger than 40, which implies that younger people don't incline to traditional Korean Mitbanchan. The dining-out motive was different in each age group. Now, the dining out motive is not restricted to home meal replacement. Social gatherings are increasing and the consumers of dining-out industry are being diversified. These suggest the increased need for classifying and analyzing the consumers by age groups to get more information on consumer behavior and tastes.

• KDI Journal of Economic Policy
• /
• v.31 no.1
• /
• pp.71-103
• /
• 2009

Considering the expenditure size of public institutions centering on public enterprises, about 28% of Korea's GDP in 2007, public institutions have significant influence on the Korean economy. However, still in the new government, there are voices of criticism about the need of constant reform on public enterprises due to their irresponsible management impeding national competitiveness. Especially, political controversy over appointment of executives such as CEOs of public enterprises has caused the distrust of the people. As one of various reform measures for public enterprises, this study analyzes the effect of internal governance structure of public enterprises on their managerial performance, since, regardless of privatization of public enterprises, improving the governance structure of public enterprises is a matter of great importance. There are only a few prior researches focusing on the governance structure and managerial performance of public enterprises compared to those of private enterprises. Most of prior researches studied the relationship between parachuting employment of CEO and managerial performance, and concluded that parachuting produces negative effect on managerial performance. However, different from the results of such researches, recent studies suggest that there is no relationship between employment type of CEOs and managerial performance in public enterprises. This study is distinguished from prior researches in view of following. First, prior researches focused on the relationship between employment type of public enterprises' CEOs and managerial performance. However, in addition to this, this study analyzes the relationship of internal auditors and managerial performance. Second, unlike prior researches studying the relationship between employment type of public corporations' CEOs and managerial performance with an emphasis on parachuting employment, this study researches impact of employment type as well as expertise of CEOs and internal auditors on managerial performance. Third, prior researchers mainly used non-financial indicators from various samples. However, this study eliminated subjectivity of researchers by analyzing public enterprises designated by the government and their financial statements, which were externally audited and inspected. In this study, regression analysis is applied in analyzing the relationship of independence and expertise of public enterprises' CEOs and internal auditors and managerial performance in the same year. Financial information from 2003 to 2007 of 24 public enterprises, which are designated by the government, and their personnel information from the board of directors are used as samples. Independence of CEOs is identified by dividing CEOs into persons from the same public enterprise and persons from other organization, and independence of internal auditors is determined by classifying them into two groups, people from academic field, economic world, and civic groups, and people from political community, government ministries, and military. Also, expertise of CEOs and internal auditors is divided into business expertise and financial expertise. As control variables, this study applied foundation year, asset size, government subsidies as a proportion to corporate earnings, and dummy variables by year. Analysis showed that there is significantly positive relationship between independence and financial expertise of internal auditors and managerial performance. In addition, although business expertise and financial expertise of CEOs were not statistically significant, they have positive relationship with managerial performance. However, unlike a general idea, independence of CEOs is not statistically significant, but it is negatively related to managerial performance. Contrary to general concerns, it seems that the impact of independence of public enterprises' CEOs on managerial performance has slightly decreased. Instead, it explains that expertise of public enterprises' CEOs and internal auditors plays more important role in managerial performance rather than their independence. Meanwhile, there are limitations in this study as follows. First, in contrast to private enterprises, public enterprises simultaneously pursue publicness and entrepreneurship. However, this study focuses on entrepreneurship, excluding considerations on publicness of public enterprises. Second, public enterprises in this study are limited to those in the central government. Accordingly, it should be carefully considered when the result of this study is applied to public enterprises in local governments. Finally, this study excludes factors related to transparency and democracy issues which are raised in appointment process of executives of public enterprises, as it may cause the issue of subjectivity of researchers.

• Journal of Intelligence and Information Systems
• /
• v.18 no.3
• /
• pp.53-77
• /
• 2012

This study analyses the difference of contents and tones of arguments among three Korean major newspapers, the Kyunghyang Shinmoon, the HanKyoreh, and the Dong-A Ilbo. It is commonly accepted that newspapers in Korea explicitly deliver their own tone of arguments when they talk about some sensitive issues and topics. It could be controversial if readers of newspapers read the news without being aware of the type of tones of arguments because the contents and the tones of arguments can affect readers easily. Thus it is very desirable to have a new tool that can inform the readers of what tone of argument a newspaper has. This study presents the results of clustering and classification techniques as part of text mining analysis. We focus on six main subjects such as Culture, Politics, International, Editorial-opinion, Eco-business and National issues in newspapers, and attempt to identify differences and similarities among the newspapers. The basic unit of text mining analysis is a paragraph of news articles. This study uses a keyword-network analysis tool and visualizes relationships among keywords to make it easier to see the differences. Newspaper articles were gathered from KINDS, the Korean integrated news database system. KINDS preserves news articles of the Kyunghyang Shinmun, the HanKyoreh and the Dong-A Ilbo and these are open to the public. This study used these three Korean major newspapers from KINDS. About 3,030 articles from 2008 to 2012 were used. International, national issues and politics sections were gathered with some specific issues. The International section was collected with the keyword of 'Nuclear weapon of North Korea.' The National issues section was collected with the keyword of '4-major-river.' The Politics section was collected with the keyword of 'Tonghap-Jinbo Dang.' All of the articles from April 2012 to May 2012 of Eco-business, Culture and Editorial-opinion sections were also collected. All of the collected data were handled and edited into paragraphs. We got rid of stop-words using the Lucene Korean Module. We calculated keyword co-occurrence counts from the paired co-occurrence list of keywords in a paragraph. We made a co-occurrence matrix from the list. Once the co-occurrence matrix was built, we used the Cosine coefficient matrix as input for PFNet(Pathfinder Network). In order to analyze these three newspapers and find out the significant keywords in each paper, we analyzed the list of 10 highest frequency keywords and keyword-networks of 20 highest ranking frequency keywords to closely examine the relationships and show the detailed network map among keywords. We used NodeXL software to visualize the PFNet. After drawing all the networks, we compared the results with the classification results. Classification was firstly handled to identify how the tone of argument of a newspaper is different from others. Then, to analyze tones of arguments, all the paragraphs were divided into two types of tones, Positive tone and Negative tone. To identify and classify all of the tones of paragraphs and articles we had collected, supervised learning technique was used. The Na$\ddot{i}$ve Bayesian classifier algorithm provided in the MALLET package was used to classify all the paragraphs in articles. After classification, Precision, Recall and F-value were used to evaluate the results of classification. Based on the results of this study, three subjects such as Culture, Eco-business and Politics showed some differences in contents and tones of arguments among these three newspapers. In addition, for the National issues, tones of arguments on 4-major-rivers project were different from each other. It seems three newspapers have their own specific tone of argument in those sections. And keyword-networks showed different shapes with each other in the same period in the same section. It means that frequently appeared keywords in articles are different and their contents are comprised with different keywords. And the Positive-Negative classification showed the possibility of classifying newspapers' tones of arguments compared to others. These results indicate that the approach in this study is promising to be extended as a new tool to identify the different tones of arguments of newspapers.

• Journal of Intelligence and Information Systems
• /
• v.19 no.4
• /
• pp.123-132
• /
• 2013

As the smartphones are equipped with various sensors such as the accelerometer, GPS, gravity sensor, gyros, ambient light sensor, proximity sensor, and so on, there have been many research works on making use of these sensors to create valuable applications. Human activity recognition is one such application that is motivated by various welfare applications such as the support for the elderly, measurement of calorie consumption, analysis of lifestyles, analysis of exercise patterns, and so on. One of the challenges faced when using the smartphone sensors for activity recognition is that the number of sensors used should be minimized to save the battery power. When the number of sensors used are restricted, it is difficult to realize a highly accurate activity recognizer or a classifier because it is hard to distinguish between subtly different activities relying on only limited information. The difficulty gets especially severe when the number of different activity classes to be distinguished is very large. In this paper, we show that a fairly accurate classifier can be built that can distinguish ten different activities by using only a single sensor data, i.e., the smartphone accelerometer data. The approach that we take to dealing with this ten-class problem is to use the ensemble of nested dichotomy (END) method that transforms a multi-class problem into multiple two-class problems. END builds a committee of binary classifiers in a nested fashion using a binary tree. At the root of the binary tree, the set of all the classes are split into two subsets of classes by using a binary classifier. At a child node of the tree, a subset of classes is again split into two smaller subsets by using another binary classifier. Continuing in this way, we can obtain a binary tree where each leaf node contains a single class. This binary tree can be viewed as a nested dichotomy that can make multi-class predictions. Depending on how a set of classes are split into two subsets at each node, the final tree that we obtain can be different. Since there can be some classes that are correlated, a particular tree may perform better than the others. However, we can hardly identify the best tree without deep domain knowledge. The END method copes with this problem by building multiple dichotomy trees randomly during learning, and then combining the predictions made by each tree during classification. The END method is generally known to perform well even when the base learner is unable to model complex decision boundaries As the base classifier at each node of the dichotomy, we have used another ensemble classifier called the random forest. A random forest is built by repeatedly generating a decision tree each time with a different random subset of features using a bootstrap sample. By combining bagging with random feature subset selection, a random forest enjoys the advantage of having more diverse ensemble members than a simple bagging. As an overall result, our ensemble of nested dichotomy can actually be seen as a committee of committees of decision trees that can deal with a multi-class problem with high accuracy. The ten classes of activities that we distinguish in this paper are 'Sitting', 'Standing', 'Walking', 'Running', 'Walking Uphill', 'Walking Downhill', 'Running Uphill', 'Running Downhill', 'Falling', and 'Hobbling'. The features used for classifying these activities include not only the magnitude of acceleration vector at each time point but also the maximum, the minimum, and the standard deviation of vector magnitude within a time window of the last 2 seconds, etc. For experiments to compare the performance of END with those of other methods, the accelerometer data has been collected at every 0.1 second for 2 minutes for each activity from 5 volunteers. Among these 5,900 ($=5{\times}(60{\times}2-2)/0.1$) data collected for each activity (the data for the first 2 seconds are trashed because they do not have time window data), 4,700 have been used for training and the rest for testing. Although 'Walking Uphill' is often confused with some other similar activities, END has been found to classify all of the ten activities with a fairly high accuracy of 98.4%. On the other hand, the accuracies achieved by a decision tree, a k-nearest neighbor, and a one-versus-rest support vector machine have been observed as 97.6%, 96.5%, and 97.6%, respectively.