• Journal of Intelligence and Information Systems
• /
• v.28 no.1
• /
• pp.263-286
• /
• 2022

This study acquires NBA statistical information for a total of 32 years from 1990 to 2022 using web crawling, observes variables of interest through exploratory data analysis, and generates related derived variables. Unused variables were removed through a purification process on the input data, and correlation analysis, t-test, and ANOVA were performed on the remaining variables. For the variable of interest, the difference in the mean between the groups that advanced to the playoffs and did not advance to the playoffs was tested, and then to compensate for this, the average difference between the three groups (higher/middle/lower) based on ranking was reconfirmed. Of the input data, only this year's season data was used as a test set, and 5-fold cross-validation was performed by dividing the training set and the validation set for model training. The overfitting problem was solved by comparing the cross-validation result and the final analysis result using the test set to confirm that there was no difference in the performance matrix. Because the quality level of the raw data is high and the statistical assumptions are satisfied, most of the models showed good results despite the small data set. This study not only predicts NBA game results or classifies whether or not to advance to the playoffs using machine learning, but also examines whether the variables of interest are included in the major variables with high importance by understanding the importance of input attribute. Through the visualization of SHAP value, it was possible to overcome the limitation that could not be interpreted only with the result of feature importance, and to compensate for the lack of consistency in the importance calculation in the process of entering/removing variables. It was found that a number of variables related to three points and errors classified as subjects of interest in this study were included in the major variables affecting advancing to the playoffs in the NBA. Although this study is similar in that it includes topics such as match results, playoffs, and championship predictions, which have been dealt with in the existing sports data analysis field, and comparatively analyzed several machine learning models for analysis, there is a difference in that the interest features are set in advance and statistically verified, so that it is compared with the machine learning analysis result. Also, it was differentiated from existing studies by presenting explanatory visualization results using SHAP, one of the XAI models.

• Journal of Preventive Medicine and Public Health
• /
• v.25 no.3 s.39
• /
• pp.247-268
• /
• 1992

This study was conducted to evaluate the effect of VDT work on eyes and vision among workers in a TV manufacturing plant. The study subjects consisted of 264 screen workers and 74 non-screen workers who were less than 40 years old male and had no history of opthalmic diseases such as corneal opacities, trauma, keratitis, etc and whose visual acuity on pre-employment health examination by Han's test chart was 1.0 or above. The screen workers were divided into two groups by actual time for screen work in a day : Group I, 60 workers, lesser than 4 hours a day and group II, 204 workers, more than 4 hours a day. From July to October 1992 a questionnaire was administered to all the study subjects for the general charateristics and subjective eye symptoms after which the opthalmologic tests such as visual acuity, spherical equivalent, lacrimal function, ocular pressure, slit lamp test, fundoscopy were conducted by one opthalmologist. The proportion of workers whose present visual acuity was decreased more than 0.15 in comparison with that on the pre-employment health examination by Han's test chart was 20.6% in Group II. 15.0% in Group I and 14.9% in non-screen workers. However, the differences in proportion were not statistically significant. The proportion of workers with decreased visual acuity was not associated with the age, working duration, use of magnifying glass and type of shift work (independent variables) in all of the three groups. However, screen workers working under poor illumination had a higher proportion of persons with decreased visual acuity than those working under adequate illumination (P<0.05) . The proportion of workers whose near vision was decreased was 27.5% in Group II, 18.3% in Group I, and 28.4% in non-screen workers and these differences in proportion were not statistically significant. Changes of near vision were not associated with 4 independent variables in all of the three groups. Six out of seven subjective eye symptoms except tearing were more common in Group I than in non-screen workers and more common in Group II than in Group I (P<0.01). Mean of the total scores for seven subjective symptoms of each worker(2 points for always, 1 point for sometimes, 0 point for never) was not significantly different between workers with decreased visual acuity and workers with no vision change. However, mean of the total scores for Group II was higher than those for the Group I and non-screen workers (P<0.01). Total eye symptom scores were significantly correlated with the grade of screen work, use of magnifying glass, and type of shift work. There was no independent variable which was correlated with the difference in visual acuity between the pre-employment health examination and the present state, the difference between far and near visions, lacrimal function, ocular pressure, and spherical equivalent. Multiple linear regression analysis for the subjective eye symptom scores revealed a positive linear relationship with actual time for screen work and shift work(P<0.01). In this study it was not observed that the VDT work decreased visual acuity but it induces subjective eye symptoms such as eye fatigue, blurred vision, ocular discomfort, etc. Maintenance of adequate illumination in the work place and control of excessive VDT work are recommended to prevent such eye symptoms.

• Journal of Intelligence and Information Systems
• /
• v.19 no.3
• /
• pp.113-125
• /
• 2013

Every company wants to know customer's requirement and makes an effort to meet them. Cause that, communication between customer and company became core competition of business and that important is increasing continuously. There are several strategies to find customer's needs, but VOC (Voice of customer) is one of most powerful communication tools and VOC gathering by several channels as telephone, post, e-mail, website and so on is so meaningful. So, almost company is gathering VOC and operating VOC system. VOC is important not only to business organization but also public organization such as government, education institute, and medical center that should drive up public service quality and customer satisfaction. Accordingly, they make a VOC gathering and analyzing System and then use for making a new product and service, and upgrade. In recent years, innovations in internet and ICT have made diverse channels such as SNS, mobile, website and call-center to collect VOC data. Although a lot of VOC data is collected through diverse channel, the proper utilization is still difficult. It is because the VOC data is made of very emotional contents by voice or text of informal style and the volume of the VOC data are so big. These unstructured big data make a difficult to store and analyze for use by human. So that, the organization need to automatic collecting, storing, classifying and analyzing system for unstructured big VOC data. This study propose an intelligent VOC analyzing system based on opinion mining to classify the unstructured VOC data automatically and determine the polarity as well as the type of VOC. And then, the basis of the VOC opinion analyzing system, called domain-oriented sentiment dictionary is created and corresponding stages are presented in detail. The experiment is conducted with 4,300 VOC data collected from a medical website to measure the effectiveness of the proposed system and utilized them to develop the sensitive data dictionary by determining the special sentiment vocabulary and their polarity value in a medical domain. Through the experiment, it comes out that positive terms such as "칭찬, 친절함, 감사, 무사히, 잘해, 감동, 미소" have high positive opinion value, and negative terms such as "퉁명, 뭡니까, 말하더군요, 무시하는" have strong negative opinion. These terms are in general use and the experiment result seems to be a high probability of opinion polarity. Furthermore, the accuracy of proposed VOC classification model has been compared and the highest classification accuracy of 77.8% is conformed at threshold with -0.50 of opinion classification of VOC. Through the proposed intelligent VOC analyzing system, the real time opinion classification and response priority of VOC can be predicted. Ultimately the positive effectiveness is expected to catch the customer complains at early stage and deal with it quickly with the lower number of staff to operate the VOC system. It can be made available human resource and time of customer service part. Above all, this study is new try to automatic analyzing the unstructured VOC data using opinion mining, and shows that the system could be used as variable to classify the positive or negative polarity of VOC opinion. It is expected to suggest practical framework of the VOC analysis to diverse use and the model can be used as real VOC analyzing system if it is implemented as system. Despite experiment results and expectation, this study has several limits. First of all, the sample data is only collected from a hospital web-site. It means that the sentimental dictionary made by sample data can be lean too much towards on that hospital and web-site. Therefore, next research has to take several channels such as call-center and SNS, and other domain like government, financial company, and education institute.

• Journal of Intelligence and Information Systems
• /
• v.23 no.4
• /
• pp.147-168
• /
• 2017

There have been many studies on accurate stock market forecasting in academia for a long time, and now there are also various forecasting models using various techniques. Recently, many attempts have been made to predict the stock index using various machine learning methods including Deep Learning. Although the fundamental analysis and the technical analysis method are used for the analysis of the traditional stock investment transaction, the technical analysis method is more useful for the application of the short-term transaction prediction or statistical and mathematical techniques. Most of the studies that have been conducted using these technical indicators have studied the model of predicting stock prices by binary classification - rising or falling - of stock market fluctuations in the future market (usually next trading day). However, it is also true that this binary classification has many unfavorable aspects in predicting trends, identifying trading signals, or signaling portfolio rebalancing. In this study, we try to predict the stock index by expanding the stock index trend (upward trend, boxed, downward trend) to the multiple classification system in the existing binary index method. In order to solve this multi-classification problem, a technique such as Multinomial Logistic Regression Analysis (MLOGIT), Multiple Discriminant Analysis (MDA) or Artificial Neural Networks (ANN) we propose an optimization model using Genetic Algorithm as a wrapper for improving the performance of this model using Multi-classification Support Vector Machines (MSVM), which has proved to be superior in prediction performance. In particular, the proposed model named GA-MSVM is designed to maximize model performance by optimizing not only the kernel function parameters of MSVM, but also the optimal selection of input variables (feature selection) as well as instance selection. In order to verify the performance of the proposed model, we applied the proposed method to the real data. The results show that the proposed method is more effective than the conventional multivariate SVM, which has been known to show the best prediction performance up to now, as well as existing artificial intelligence / data mining techniques such as MDA, MLOGIT, CBR, and it is confirmed that the prediction performance is better than this. Especially, it has been confirmed that the 'instance selection' plays a very important role in predicting the stock index trend, and it is confirmed that the improvement effect of the model is more important than other factors. To verify the usefulness of GA-MSVM, we applied it to Korea's real KOSPI200 stock index trend forecast. Our research is primarily aimed at predicting trend segments to capture signal acquisition or short-term trend transition points. The experimental data set includes technical indicators such as the price and volatility index (2004 ~ 2017) and macroeconomic data (interest rate, exchange rate, S&P 500, etc.) of KOSPI200 stock index in Korea. Using a variety of statistical methods including one-way ANOVA and stepwise MDA, 15 indicators were selected as candidate independent variables. The dependent variable, trend classification, was classified into three states: 1 (upward trend), 0 (boxed), and -1 (downward trend). 70% of the total data for each class was used for training and the remaining 30% was used for verifying. To verify the performance of the proposed model, several comparative model experiments such as MDA, MLOGIT, CBR, ANN and MSVM were conducted. MSVM has adopted the One-Against-One (OAO) approach, which is known as the most accurate approach among the various MSVM approaches. Although there are some limitations, the final experimental results demonstrate that the proposed model, GA-MSVM, performs at a significantly higher level than all comparative models.

• Asia-Pacific Journal of Business Venturing and Entrepreneurship
• /
• v.12 no.5
• /
• pp.127-139
• /
• 2017

In recent 10 years the attention to social entrepreneurship has raised increasing among scholars, public sector, and community development. However less research has been conducted on how social entrepreneurship intention create a social enterprise and what factors can be affected to the social entrepreneurial intentions. This paper aims at contributing to identify the antecedents of entrepreneurial behavior and intentions. Especially, we have had a strong interests in compassion factors which haven't been used as important variables to encourage for people to do social entrepreneurial activities. Also, we try to find the moral obligation and perceived social support as antecedents of social entrepreneurial intentions. Finding show that compassion and moral obligation affect to the social entrepreneurial intention. Especially this study identify the external factor of society with the variable, perceived social support. Once individuals recognize that the infrastructure and societal positive mood on social entrepreneurship is friendly to social entrepreneurship, people have a tendency to try to do some social entrepreneurial activities. Only few empirical studies exist in this research domain. A study of more than 271 Korean college students has studied which personal traits predict certain characteristics of social entrepreneurs (such as having social vision or looking for social innovational opportunities). In addition to those antecedents, students experience is the critical factor that enabled continued expansion of the social entrepreneurial activities. The results of this research show how we can nurture social entrepreneurs and how we can develop the social environment to promote social entrepreneurship.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.45 no.3
• /
• pp.21-29
• /
• 2017

In this study of visitor place attachment to parks, Scannel and Gilford considered the dimensions of place attachment, and proposed the theoretical framework of PPP(Place, Person, Process) that should be systematically studied from the perspective of person, place and psychological processes. According to Scannel's theoretical basis, this paper puts forward the hypothesis of a structure model of place attachment. In the model, the five independent variables of people, places, cognition, emotion, and behavior have influenced the dependent variable of place attachment. The questionnaire was conducted on 18 neighborhood parks in Kunming, China, and the residents of the 5 neighborhood parks in Daejeon, South Korea. A total of 1,645 valid samples of the questionnaire survey were collected. Through confirmatory factor analysis(CFA) results of the inspection of the various elements, it was shown that the reliabilities of 6 latent variables, such as people, places, cognition, affection, behavior and place attachment, which were composed with the observed variables(30 observed variables in Daejeon, 19 observed variables in Kunming), were all above 0.7 and the data were fit for this study. The hypothesis test results found that the physical environment of the neighborhood parks such as a comfortable environment, pleasant road and convenient facilities would increase the rate of visitors coming back. From the park management perspective, to increase the amount of visitors to the park, the park should increase visibility, provide more organized, varied activities and meetings, or special exhibitions according to the particular characteristics of the individual park, to increase awareness of the park. From the park visitors' psychological perspective, visitors are seeking to enjoy the park facilities and environment not only to bring physical relaxation, but also to bring about a psychological cure. With the commonality of attachment structure between the two countries, to improve the place attachment of neighborhood park visitors, collecting regularly visitor feedback will facilitate the sustainable development of neighborhood park attachment.

• Journal of Physiology & Pathology in Korean Medicine
• /
• v.17 no.5
• /
• pp.1157-1176
• /
• 2003

The following conclusions are induced from a study on the acupuncture therapy depending on hyungsang of the persons. The study is made on the basis of 'Internal classic (內經)& and &Clinical Lectures by Dr. Jeesan&. The acupuncture originated from the treatment of spasm with numbness in the southern area. The acupuncture is basically a remedy for the exterior disease of meridian but also it can be a cure for the interior disease of Jang and obstinate disease with accurate method. Three mechanisms of acupuncture are described in 'Internal classic'. The first is to make meridian circulate smoothly. The second is to regulate Ki and Hyul. The third is to regulate points through which the meridian-Ki goes in and out smoothly or adversely. There are two ways of acupuncture in 'Internal classic'. One is based on pulse and symptom and the other on the Hyungsang. The former is more generally used therapy, to which depletion method, Asi point method(阿是穴 療法), Inyoung-kigu pulse comparison method (人迎氣口脈法) and method depending on jang-bu disease belong. Acupuncture is done on Su points(輸穴) and back-su point(背兪穴) in case of jang-disease. In case of bu-disease, the treatment is done on Hap points(合穴) and Mo-points(募穴). The latter includes two methods; one according to invariable Hyungsang. And the other to variable Hyungsang. The method of acupuncture according to invariable Hyungsang usually selects Won-points(原穴). Different Hyungsang requires different method of acupuncture; In case of Dam type, the acupuncture is mainly practiced on four-Kwan points with reinforcing and reducing methods achieved by the direction of the needle tip pointing to. In case of Bangkwang type, the acupuncture is usually done on Jungwan(中脘) and Poongyung(豊隆) with reinforcing and reducing methods by means of respiration. In case of female, more effective are the acupoints on the right and lateral parts of the body selected on the basis of five su-points of the twelve meridians matching the heavenly stems and earthly branches. In case of male, more effective are the acupoints on the left, front and rear parts of the eight extra meridians. In case of acupuncture to the person with Hyungsang of five jang and six bu, each person's intrinsic Hyung, color, pulse, must be observed. Because symptoms of jang-bu disease also must be checked up. Acupuncture is done on the Won-points of the meridians related to the jang and bu where the disease starts. The disease of five jang is so obstinate that it requires both of medication and acupuncture for a long time. In case of acupuncture to the person with Hyungsang of animal types, diagnosis is made on the basis of shape, temper, function and color. And the treatment is given on the Won-points of corresponding exterior and interior meridians. For the fish type, the acupuncture is done on the kidney meridian of foot-soyin and the urinary bladder of foot-taiyang. For the bird type, on the heart meridian of hand-soyin, the pericardium meridian of hand-gualyin, and the small intestine meridian of hand-taiyang For the deer type, on the liver meridian of foot-gualyin and the gallbladder meridian of foot-soyang. For the turtle type, on the lung meridian of hand-taiyin and the large intestine meridian of hand-yangmyung.

• Journal of Periodontal and Implant Science
• /
• v.24 no.1
• /
• pp.131-143
• /
• 1994

For maintenance of exposed implant in healthy state, it is necessary to treat the surface of implant fixture and provide the surface adjustable to surrounding tissues. Variable techniques have been introduced such as citric acid and air-abrasive system to treat the failed implant. Although when the rough surface of HA coated implant was exposed to oral environment, the surface treatment method with citric acid or air-abrasive system is effective for removal of bacterial endotoxin, it is unsuccessful to prevent plaque deposition due to difficulty in removal of rough surface of HA coated implant. Thus, in this study the method that removes bacterial endotoxin and makes smooch surface without alteration of surface characteristics was studied. HA coated disc manufactured by IMZ Co. Was treated with high speed diamond bur, low speed diamond bur, stone bur, rubber point, jetpolisher. And then its surface state was examined with profilometer and SEM to evaluate the surface smoothness, and its surface component was analyzed with EDX to evaluate wheter the surface characteristics were altered or not. As a result, following results were obtained. When the surface roughness of each implant disc was measured by profilometer, the group I showed a $R_{max}\;2.11{\mu}m$ and the group II, III, IV, V showed a $R_{max2}\;4.17{\mu}m$, $7.28{\mu}m$, $8.61{\mu}m$ and $39.44{\mu}m$ respectively. That is, surface smoothness was highest in the group I and it has been gradually decreased in the group II, III, IV and V. Under the SEM examination, the group I showed relatively smooth surface and the group II showed slightly rougher surface than the group I due to partially remaining HA particles while most HA particle was removed. The group III and IV showed rough topography due to HA particles that was not grinded, and HA coated surface in group V showed very irregular surface with deep groove and prominence. In cross-sectional view, the group I showed uniform surface, and the group III, IV showed rough surface due to remaining HA particles but the thickness of HA coating was remarkably reduced. The group II has similar pattern in group I, and the group V showed about $40{\mu}m$ thickness although it was not constant. By analysis of surface component with EDX, the group II in which the grinding was effective showed a small quantity of calcium and phosphorous and the group III, IV, in which the grinding was incomplete showed calcium and phosphorus peak. In all experimental group, no other than titanium, aluminum, calcium, phosphorus was observed.

• Korean Journal of Optics and Photonics
• /
• v.21 no.2
• /
• pp.82-88
• /
• 2010

Photopolymer is a material for recording three dimensional holograms containing photo information. Photopolymer has been found to be a proper material due to many advantages such as high DE value, easy processing, and low price. Compositions of PVA, monomer, initiater and photosensitizer were determined by previous experiments and the compositions of $SeO_2$ and $TiO_2$ were considered as variable to find out the effects of $TiO_2$ on DE. The DE values were constant for the varying compositions of $TiO_2$ (0.1 mg~1.0 mg). In other words, $TiO_2$ is not directly effective on the DE values. Composition change experiments from $SeO_2$ 0.1 mg, $TiO_2$ 0.9 mg to $SeO_2$ 0.9 mg, $TiO_2$ 0.1 showed a maximum DE value of 73.75% at a component of $SeO_2$ 0.8 mg, $TiO_2$ 0.2 mg. It seemed that regardless of the amount of $TiO_2$, increasing the amount of $SeO_2$ gently increases DE`s. If nano particles are heavily added, transparent films could not be made due to the separation of particles by the solubility decrease. Photopolymer films could be made with high DE values for an extensive angle range if $TiO_2$ additions were kept minimum and $SeO_2$ additions were kept maximum.

• Management & Information Systems Review
• /
• v.32 no.5
• /
• pp.193-231
• /
• 2013

As developed IT, Word-of-Mouth(WOM) used varied terms as buzz marketing and viral marketing, and impressed that importance. Despite introduced new marketing tool on managers and professionals, online word-of-mouth including SNS lack of study on social network what based viral in marketing. In social network, patterns of relationship between individuals influence each other individual behaviors. Therefore this research grouped friendship-network and advice-network by characteristics, studied on trust of information source that antecedents of word-of-mouth in network. This study examined that affect- and cognition based trust affect WOM acceptance as WOM behaviors and examined effect of type of product as moderating variable. Additional this literature studied that WOM acceptance affect WOM recommend. To find the Influence of Trust on Word-of-Mouth Behaviors, a survey has done 206 samples(undergraduate students). The results of this study are as following : First, type of trust different friendship network and advice network. Affect-based trust is outstanding in friendship network than in advice network, while cognition-based trust stands out in advice network than another. Second, affect- and cognition based trust positive affect WOM acceptance. Contrary to expectations, what is preconceived trust in network have a similar effect for WOM acceptance regardless of type of trust. Third, WOM acceptance positive affect WOM recommend. Fourth, affect based trust affect WOM acceptance of hedonic product rather than utilitarian product. Upon especially in friendship network terms, affect-based trust has a more effect on WOM acceptance than cognition-based trust. This study has many implications. First, it is important that trust what have an influence WOM acceptance grouped affect- and cognition based trust. Second, it confirmed that trust is antecedents of positive WOM. Third, it is important that network grouped friendship network and advice-network by trust. Fourth, it gave managerial implications that they have to supply WOM through which network by type of product. We This study classified network and trust based on previous study. Then it examined relations between WOM behaviors. Further research could do enrich various things for example various age group, valence of message, quality of information.