• Journal of Intelligence and Information Systems
• /
• v.21 no.4
• /
• pp.1-16
• /
• 2015

Traffic accident is one of the major cause of death worldwide for the last several decades. According to the statistics of world health organization, approximately 1.24 million deaths occurred on the world's roads in 2010. In order to reduce future traffic accident, multipronged approaches have been adopted including traffic regulations, injury-reducing technologies, driving training program and so on. Records on traffic accidents are generated and maintained for this purpose. To make these records meaningful and effective, it is necessary to analyze relationship between traffic accident and related factors including vehicle design, road design, weather, driver behavior etc. Insight derived from these analysis can be used for accident prevention approaches. Traffic accident data mining is an activity to find useful knowledges about such relationship that is not well-known and user may interested in it. Many studies about mining accident data have been reported over the past two decades. Most of studies mainly focused on predict risk of accident using accident related factors. Supervised learning methods like decision tree, logistic regression, k-nearest neighbor, neural network are used for these prediction. However, derived prediction model from these algorithms are too complex to understand for human itself because the main purpose of these algorithms are prediction, not explanation of the data. Some of studies use unsupervised clustering algorithm to dividing the data into several groups, but derived group itself is still not easy to understand for human, so it is necessary to do some additional analytic works. Rule based learning methods are adequate when we want to derive comprehensive form of knowledge about the target domain. It derives a set of if-then rules that represent relationship between the target feature with other features. Rules are fairly easy for human to understand its meaning therefore it can help provide insight and comprehensible results for human. Association rule learning methods and subgroup discovery methods are representing rule based learning methods for descriptive task. These two algorithms have been used in a wide range of area from transaction analysis, accident data analysis, detection of statistically significant patient risk groups, discovering key person in social communities and so on. We use both the association rule learning method and the subgroup discovery method to discover useful patterns from a traffic accident dataset consisting of many features including profile of driver, location of accident, types of accident, information of vehicle, violation of regulation and so on. The association rule learning method, which is one of the unsupervised learning methods, searches for frequent item sets from the data and translates them into rules. In contrast, the subgroup discovery method is a kind of supervised learning method that discovers rules of user specified concepts satisfying certain degree of generality and unusualness. Depending on what aspect of the data we are focusing our attention to, we may combine different multiple relevant features of interest to make a synthetic target feature, and give it to the rule learning algorithms. After a set of rules is derived, some postprocessing steps are taken to make the ruleset more compact and easier to understand by removing some uninteresting or redundant rules. We conducted a set of experiments of mining our traffic accident data in both unsupervised mode and supervised mode for comparison of these rule based learning algorithms. Experiments with the traffic accident data reveals that the association rule learning, in its pure unsupervised mode, can discover some hidden relationship among the features. Under supervised learning setting with combinatorial target feature, however, the subgroup discovery method finds good rules much more easily than the association rule learning method that requires a lot of efforts to tune the parameters.

• Journal of Intelligence and Information Systems
• /
• v.24 no.4
• /
• pp.197-217
• /
• 2018

Music is one of the most creative act that can express human sentiment with sound. Also, since music invoke people's sentiment to get empathized with it easily, it can either encourage or discourage people's sentiment with music what they are listening. Thus, sentiment is the primary factor when it comes to searching or recommending music to people. Regard to the music recommendation system, there are still lack of recommendation systems that are based on customer sentiment. An algorithm's that were used in previous music recommendation systems are mostly user based, for example, user's play history and playlists etc. Based on play history or playlists between multiple users, distance between music were calculated refer to basic information such as genre, singer, beat etc. It can filter out similar music to the users as a recommendation system. However those methodology have limitations like filter bubble. For example, if user listen to rock music only, it would be hard to get hip-hop or R&B music which have similar sentiment as a recommendation. In this study, we have focused on sentiment of music itself, and finally developed methodology of defining new index for music recommendation system. Concretely, we are proposing "SWEMS" index and using this index, we also extracted "Sentiment Pattern" for each music which was used for this research. Using this "SWEMS" index and "Sentiment Pattern", we expect that it can be used for a variety of purposes not only the music recommendation system but also as an algorithm which used for buildup predicting model etc. In this study, we had to develop the music recommendation system based on emotional adjectives which people generally feel when they listening to music. For that reason, it was necessary to collect a large amount of emotional adjectives as we can. Emotional adjectives were collected via previous study which is related to them. Also more emotional adjectives has collected via social metrics and qualitative interview. Finally, we could collect 134 individual adjectives. Through several steps, the collected adjectives were selected as the final 60 adjectives. Based on the final adjectives, music survey has taken as each item to evaluated the sentiment of a song. Surveys were taken by expert panels who like to listen to music. During the survey, all survey questions were based on emotional adjectives, no other information were collected. The music which evaluated from the previous step is divided into popular and unpopular songs, and the most relevant variables were derived from the popularity of music. The derived variables were reclassified through factor analysis and assigned a weight to the adjectives which belongs to the factor. We define the extracted factors as "SWEMS" index, which describes sentiment score of music in numeric value. In this study, we attempted to apply Case Based Reasoning method to implement an algorithm. Compare to other methodology, we used Case Based Reasoning because it shows similar problem solving method as what human do. Using "SWEMS" index of each music, an algorithm will be implemented based on the Euclidean distance to recommend a song similar to the emotion value which given by the factor for each music. Also, using "SWEMS" index, we can also draw "Sentiment Pattern" for each song. In this study, we found that the song which gives a similar emotion shows similar "Sentiment Pattern" each other. Through "Sentiment Pattern", we could also suggest a new group of music, which is different from the previous format of genre. This research would help people to quantify qualitative data. Also the algorithms can be used to quantify the content itself, which would help users to search the similar content more quickly.

• The Journal of Korean Academy of Prosthodontics
• /
• v.52 no.4
• /
• pp.312-316
• /
• 2014

Purpose: The aim of this study was to compare the fracture toughness of currently available resin cements for zirconia restorations and evaluate the effect of water storage on fracture toughness of those resin cements. Materials and methods: Single-edge notched specimens ($3mm{\times}6mm{\times}25mm$) were prepared from three currently available dual cure resin cements for zirconia restorations (Panavia F 2.0, Clearfil SA luting and Zirconite). Each resin cement was divided into four groups: immersed in distilled water at $37^{\circ}C$ for 1 (Control group), 30, 90, or 180 days (n=5). Specimens were loaded in three point bending at a cross-head speed of 0.1 mm/s. The maximum load at specimen failure was recorded and the fracture toughness ($K_{IC}$) was calculated. Data were analyzed using one-way ANOVA and multiple comparison $Scheff{\acute{e}}$ test (${\alpha}$=.05). Results: In control group, the mean $K_{IC}$ was $3.41{\pm}0.64MN{\cdot}m^{-1.5}$ for Panavia F, 2.0, $3.07{\pm}0.41MN{\cdot}m^{-1.5}$ for Zirconite, $2.58{\pm}0.30MN{\cdot}m^{-1.5}$ for Clearfil SA luting respectively, but statistical analysis revealed no significant difference between them. Although a gradual decrease of $K_{IC}$ in Panavia F 2.0 and gradual increases of KIC in Clearfil SA luting and Zirconite were observed with storage time, there were no significant differences between immersion time for each cement. Conclusion: The resin cements for zirconia restorations exhibit much higher $K_{IC}$ values than conventional resin cements. The fracture toughness of resin cement for zirconia restoration would not be affected by water storage.

• Neonatal Medicine
• /
• v.18 no.2
• /
• pp.221-227
• /
• 2011

Purpose: The objectives were to identify the characteristics of neonates with hydrops fetalis, and to identify the risk factors associated with mortality. Methods: A retrospective review of AMC (Asan Medical Center) dataset was performed from January 1990 to June 2009. The characteristics of 71 patients with hydrops fetalis were investigated and they were divided into two groups: the survived group and the expired group. Various perinatal and neonatal factors in two groups were compared to find out risk factors associated with mortality based on univariate analysis, followed by multiple regression analyses (SPSS version 18.0). Results: Of those 71 neonates (average gestational age: 33 weeks, birth weight: 2.6 kg), 38 survived, 33 died, resulting in overall mortality rate of 46.5%. The most common etiology was idiopathic followed by chylothorax, cardiac anomalies, twin-to-twin transfusion, meconium peritonitis, cardiac arrythmias, and congenital infections. Factors that were associated independently with mortality in logistic regression analyses were low 5-minutes Apgar score, hyaline membrane disease and delayed in achieving 50th percentile ideal body weight for appropriate gestational age by 10 days. Conclusion: In this study, 5-minutes Apgar score, hyaline membrane disease and delayed in achieving 50th percentile ideal body weight for appropriate gestational age by 10 days were significant risk factors associated with mortality in hydrops fetalis. Therefore, the risk of death among neonates with hydrops fetalis depends on the illness immediately after birth and severity of hydrops fetalis. Informations from this study may prove useful in prediction of prognosis to neonates with hydrops fetalis.

• Journal of Preventive Medicine and Public Health
• /
• v.31 no.3 s.62
• /
• pp.503-515
• /
• 1998

In order to investigate the factors influencing QOL and life satisfaction of the physically disabled, the author interviewed, by using structuralized questionnaire, 440 individuals among the physically disabled who were participated in Health Examination from April to July, 1997. The questionnaire consisted of the general characteristics(sex, age, marital status, family number, etc), the Reintegration to Normal Living Index(RNLI) to assess QOL, and the single item of five-likert scale to evaluate life satifaction. The means of RNLI were $16.2{\pm}4.8$ in total score, $12.2{\pm}3.4$ in daily functioning and $4.0{\pm}2.1$ in perception of self. The respondents were less reintegrated toward social activities and relationships than impairments or disabilities. While the satisfied group was 47.3%, the dissatisfied group was 52.7%. As the results of multiple regression and logistic regression analysis, the significant predictors of QOL were age, education, job, grade of disability and subjective health status. The life satisfaction were related to economic status, job and subjective health status. To improve QOL and life satisfaction of the physically disabled, it is important that we improve their basic socioeconomic status by getting a job through rehabilitation education and induce them to have positive self-assessment by extending the opportunity of social participation.

• Journal of Preventive Medicine and Public Health
• /
• v.31 no.3 s.62
• /
• pp.564-578
• /
• 1998

Most extended care facilities have admitted both the healthy and unhealthy elderly, among which members' average caring demand vary. The Aged Welfare Law, however, currently provides no reasonable basis on the staffing policy for extended care facilities. It just reflects the admitted number of the elderly rather than differences in members' average caring demand among facilities. This study is designed to estimate the need for caring staff on the basis of the correlation between the individual health status measured by various tools including Activities of Daily Living and caring demand by actual service time for each one. The sample included all of the admitted elderly(187 persons) in 4 extended care facilities, two in Seoul and the other two in Kangwon-Do over the survey period October 5 through October 20, 1996. The survey process consisted of 3 stages. (1) The current staffing information was collected through self-completed written questionnaires left for head of official in each facility. (2) Six graduate students at School of Public Health interviewed all residents to collect information on their health status and sociodemographics. The response rate for the interview was relatively high(85%). (3) Information on direct and indirect caring time consumed for each residents came from self-completed written questionnaires given to nurses and helpers in each target facilities. Analysis of the data was made using Pearson's correlation and multiple regression technique through SAS program. Based on this procedure, the following was found. 1. No facility meet the staffing standard in the Aged Welfare Law completely. 2. It is actual service time that is most correlated with ADL(Activities of daily living). 3. When all of the elderly are divided by four groups based on the level of ADL, the mean values of needed caring time in each group are 15, 21, 36 and 88 minutes respectively. 4. There is no significant difference among facilities in distribution of elderly person by group. 5. No facility meets the estimated number of nurses and helpers which reflects health status of the admitted elderly. Therefore, it is required that severity of the admitted elderly be considered in establishing staffing standard for extended care facility.

• Clinical and Experimental Pediatrics
• /
• v.48 no.1
• /
• pp.27-33
• /
• 2005

Purpose : To determine incidence, characteristics and risk factors associated with intrauterine intraventricular hemorrhage(IU-IVH) among premature infants. Methods : The medical records of infants with intraventricular hemorrhage(IVH) admitted to the neonatal intensive care unit of Asan Medical Center from January 1999 to June 2003 were reviewed retrospectively. Infants whose IVH with cystic change were detected within five days of life were defined as the IU-IVH group. The control group included those without any IVH. Various maternal and neonatal factors were evaluated between the IU-IVH and control groups, and risk factors for IU-IVH were identified using multiple logistic regression analysis. Results : The incidence of IU-IVH was 49/1024(15.9%). Mothers who are younger, primiparous, use less antenatal steroid, and neonates with greater incidence of neonatal respiratory distress syndrome, had higher incidences of IU-IVH compared to neonates with normal neurosonography. Risk factors associated with IU-IVH included neonatal respiratory distress syndrome and placenta infarct by placenta biopsy. Most infants with IU-IVH were ${\geq}1,501g$, ${\geq}34$ weeks gestational age and had low grade IVH. The size of the cysts associated with IU-IVH remained the same or disappeared in 96 %. IU-IVH does not seem to affect short-term neurodevelopmental outcome although a longer period of follow-up is needed. Conculusion : IU-IVH occurred mostly in ${\geq}1,501g$, ${\geq}34$ weeks infants with grade I IVH without developmental delays. However, the high incidence of total IVH merits more attention in terms of awareness of its existence as an unusual IVH among premature infants.

• Journal of Preventive Medicine and Public Health
• /
• v.29 no.3 s.54
• /
• pp.579-596
• /
• 1996

The first purpose of this study was to compare the mean Knowledge, Attitude & Practice(KAP) scores concerning major occupational health between two types of occupational health management in small & medium-scale textile factories. Sample number was 1,138 workers. In small & medium factories with 300 workers or less, they have two alternatives for occupational health management ; Individual Health Management or Vicarious Health Management. The second purpose was to determine if there was a relationship between duration of Vicarious health management and mean KAP scores for 694 workers from the Vicarious Health Management group. We used self-administered questionnaires for measuring KAP. The results of this study were; 1. There was significant difference concerning all socioeconomic & occupational variables (gender, age, education level, marital status, Blue/White collar, working duration of present factory, working duration of total factories) except for shift. 2. There was no significant difference of mean scores of Knowledge and Attitude between the Individual Health Management Group and the Vicarious Health Management. But, the mean score of Practice was significantly different. 3. According to multiple regression analysis, the mean score of Knowledge significantly increased with age, education level, working duration of all factories increased. The mean score of Attitude significantly increased with age, education level and married. The mean score of Practice significantly increased with age, education level, working duration of all factories, married and men. 4. In conclusion, The type of Health management did not significantly affect the mean KAP scores concerning major occupational Health problems. 5. But, the Duration of Vicarious Health management significantly increased the mean KAP scores.

• Clinical and Experimental Pediatrics
• /
• v.48 no.9
• /
• pp.991-997
• /
• 2005

Purpose : Cow's milk protein-induced enterocolitis(CMPIE) is a symptom complex of vomiting and/or diarrhea caused by delayed hypersensitivity and may result in serious complications. This study was undertaken to identify high risk factors to facilitate the early recognition of CMPIE. Methods : We reviewed the data of 101 patients, aged 15 to 45 days, admitted due to vomiting and/or diarrhea between 2003 and 2004. After excluding 13 patients absolutely breast-fed and 2 patients transferred from other hospitals with the impression of CMPIE, the 86 study subjects were divided into three groups based on the underlying etiologies; CMPIE, infectious and non-infectious group. Results : CMPIE was diagnosed in 11 patients(12.8%). On admission, failure to gain weight(P=0.003), hypoalbuminemia(P=0.003), peripheral leukocytosis(P=0.015), and metabolic acidosis(P=0.014) were more significant in the CMPIE group than in the others. Multiple logistic regression analysis showed that the independent predictors of high risks for CMPIE were failure to gain weight <10 g/day(OR, 10.25[95% CI, 1.62-65.06]) and serum hypoalbuminemia <3.5 g/dL(OR, 9.18[95% CI, 1.69-49.74]). Cow's milk challenges were performed in the 11 CMPIE patients; vomiting(81.8%), abnormal stool test(80.0%), peripheral leukocyte count and absolute neutrophil count(ANC) increase(100.0%) (P<0.05), and enteropathy(100.0%). Conclusion : CMPIE is not a rare clinical disease in early infancy. The high risk factors of CMPIE were identified as follow : failure to gain weight below 10 g/day, hypoalbuminemia on admission and a rapid decrease during admission. Cow's milk challenge test with endoscopic duodenal biopsy was helpful to confirm CMPIE.

• Korean Journal of Adult Nursing
• /
• v.12 no.1
• /
• pp.99-111
• /
• 2000

The purpose of this study was to identify risk factors related to sexual function of women with coronary artery disease, and to determine the predictors of sexual function. The study design, a descriptive correlational study, was done through structural questionnaire and interview. A total of 50 subjects from C University Hospital at Kwang-ju city who have undergone coronary angiography at department of cardiology were observed and interviewed from Feb. 22, 1999 to March. 23, 1999. The number of affected vessels, the level of total serum cholesterol, and the ejection fraction of 2-D echo cardiography were analyzed to evaluate the severity of coronary artery disease. And also type A behavior pattern, health behavior, Brief Index of Sexual Functioning for Women (BISF-W) were measured. The data obtained were analyzed using percentage, mean and standard deviation, t-test, ANOVA, Pearson's correlation coefficient, and stepwise multiple regression analysis via SPSS PC+. The results of this study were as follows: 1. The mean age of the subjects were 58.1 and 72.0% of those have been married over 30 years. Seventy two percentage were unemployed and monthly family income of 56.6% was less than 1,000,000 won (approximately $ 840). Eighty percent were in their postmenopausal state, and the frequency of sexual intercourse of 84.0% were two to three times per month. 2. The scores of type A behavior pattern were from 16 to 38(mean 24.94) and health behavior ranged from 21 to 43(mean 31.2). Abstinence from smoking, alcohol, and caffeine were best compliant factors and weight control and exercise were least abided ones. The result of 2D-ECHO EF showed that the half of the subjects were abnormal, and 24% had more than 240mg/dl of total serum cholesterol. The coronary angiography showed that 64% of the subjects had more than one affected vessels. 3. The predictors to explain the factor score of 'orgasm' were number of health examination, the pre- or post-menopausal state, protestant, number of coronary vessel affected, level of serum total cholesterol, and comorbid group of hypertension and diabetes, and it's total variance accounted for 52.4%. The predictors to explain the factor score of 'sexual activity' were comorbid group of hypertension and diabetes and type A behavior pattern, which accounted for 22.4% of total variance. The predictors to explain the factor score of 'sexual satisfaction' were type A behavior pattern, no religion, exercise, level of serum total cholesterol, and pre or post menopausal state, which accounted for 52.1%. The predictors to explain the factor score of 'sexual desire' were the period of marriage, type A behavior, employment or unemployment, and weight control, which accounted for 43.2%. The predictors to explain the factor score of 'external force of sexual functioning' were physical overload and exercise, which accounted for 41.1%. The predictors to explain the factor score of 'sexual activity' were family monthly income, catholics, and exercise, and which accounted for 35.4%. Above results lead us to some consensus that sexual function of women with coronary artery disease is related to various factors including vasogenic factors such as total serum cholesterol level, number of coronary vessel affected, an endocrinal factor such as menopausal state, and type A behavior pattern as a sociopshychological factor. And also health behaviors such as fitness care, overwork, weight control, and emotional tension are contributed to sexual function.