• Journal of Intelligence and Information Systems
• /
• v.22 no.3
• /
• pp.113-127
• /
• 2016

Among the techniques for recommendation, collaborative filtering (CF) is commonly recognized to be the most effective for implementing recommender systems. Until now, CF has been popularly studied and adopted in both academic and real-world applications. The basic idea of CF is to create recommendation results by finding correlations between users of a recommendation system. CF system compares users based on how similar they are, and recommend products to users by using other like-minded people's results of evaluation for each product. Thus, it is very important to compute evaluation similarities among users in CF because the recommendation quality depends on it. Typical CF uses user's explicit numeric ratings of items (i.e. quantitative information) when computing the similarities among users in CF. In other words, user's numeric ratings have been a sole source of user preference information in traditional CF. However, user ratings are unable to fully reflect user's actual preferences from time to time. According to several studies, users may more actively accommodate recommendation of reliable others when purchasing goods. Thus, trust relationship can be regarded as the informative source for identifying user's preference with accuracy. Under this background, we propose a new hybrid recommender system that fuses CF and social network analysis (SNA). The proposed system adopts the recommendation algorithm that additionally reflect the result analyzed by SNA. In detail, our proposed system is based on conventional memory-based CF, but it is designed to use both user's numeric ratings and trust relationship information between users when calculating user similarities. For this, our system creates and uses not only user-item rating matrix, but also user-to-user trust network. As the methods for calculating user similarity between users, we proposed two alternatives - one is algorithm calculating the degree of similarity between users by utilizing in-degree and out-degree centrality, which are the indices representing the central location in the social network. We named these approaches as 'Trust CF - All' and 'Trust CF - Conditional'. The other alternative is the algorithm reflecting a neighbor's score higher when a target user trusts the neighbor directly or indirectly. The direct or indirect trust relationship can be identified by searching trust network of users. In this study, we call this approach 'Trust CF - Search'. To validate the applicability of the proposed system, we used experimental data provided by LibRec that crawled from the entire FilmTrust website. It consists of ratings of movies and trust relationship network indicating who to trust between users. The experimental system was implemented using Microsoft Visual Basic for Applications (VBA) and UCINET 6. To examine the effectiveness of the proposed system, we compared the performance of our proposed method with one of conventional CF system. The performances of recommender system were evaluated by using average MAE (mean absolute error). The analysis results confirmed that in case of applying without conditions the in-degree centrality index of trusted network of users(i.e. Trust CF - All), the accuracy (MAE = 0.565134) was lower than conventional CF (MAE = 0.564966). And, in case of applying the in-degree centrality index only to the users with the out-degree centrality above a certain threshold value(i.e. Trust CF - Conditional), the proposed system improved the accuracy a little (MAE = 0.564909) compared to traditional CF. However, the algorithm searching based on the trusted network of users (i.e. Trust CF - Search) was found to show the best performance (MAE = 0.564846). And the result from paired samples t-test presented that Trust CF - Search outperformed conventional CF with 10% statistical significance level. Our study sheds a light on the application of user's trust relationship network information for facilitating electronic commerce by recommending proper items to users.

• Journal of Intelligence and Information Systems
• /
• v.22 no.3
• /
• pp.71-89
• /
• 2016

Sentiment analysis is used for identifying emotions or sentiments embedded in the user generated data such as customer reviews from blogs, social network services, and so on. Various research fields such as computer science and business management can take advantage of this feature to analyze customer-generated opinions. In previous studies, the star rating of a review is regarded as the same as sentiment embedded in the text. However, it does not always correspond to the sentiment polarity. Due to this supposition, previous studies have some limitations in their accuracy. To solve this issue, the present study uses a supervised sentiment classification model to measure a more accurate sentiment polarity. This study aims to propose an advanced sentiment classifier and to discover the correlation between movie reviews and box-office success. The advanced sentiment classifier is based on two supervised machine learning techniques, the Support Vector Machines (SVM) and Feedforward Neural Network (FNN). The sentiment scores of the movie reviews are measured by the sentiment classifier and are analyzed by statistical correlations between movie reviews and box-office success. Movie reviews are collected along with a star-rate. The dataset used in this study consists of 1,258,538 reviews from 175 films gathered from Naver Movie website (movie.naver.com). The results show that the proposed sentiment classifier outperforms Naive Bayes (NB) classifier as its accuracy is about 6% higher than NB. Furthermore, the results indicate that there are positive correlations between the star-rate and the number of audiences, which can be regarded as the box-office success of a movie. The study also shows that there is the mild, positive correlation between the sentiment scores estimated by the classifier and the number of audiences. To verify the applicability of the sentiment scores, an independent sample t-test was conducted. For this, the movies were divided into two groups using the average of sentiment scores. The two groups are significantly different in terms of the star-rated scores.

• Journal of Korean Orthopaedic Sports Medicine
• /
• v.5 no.1
• /
• pp.41-44
• /
• 2006

Purpose: To evaluate the short-term results of arthroscopic ACL reconstruction using fresh frozen Achilles allograft Materials & Methods: From March 2002 to March 2004, arthroscopic ACL reconstructions using fresh frozen Achilles allograft were performed in 25 knees of 25 patients. The average age at operation was 30.1 years (range, 18-50 years) and the average follow-up was 17 months (range, 12 to 27months). Preoperative and follow-up clinical results were evaluated using the Lysholm knee score, IKDC knee rating system, physical examination and KT-2000 arthrometer. Results: The Lachman test was positive in 25 patients preoperatively and 18 patients(72%) had negative results at latest follow-up. The average side-to-side differences of anterior tibial translation using KT-2000 arthrometer under loading of 301b were improved from $7.9{\pm}2.4mm\;to\;2.6{\pm}1.6mm$. The average Lysholm score was improved from $61.1{\pm}13.9\;to\;93.5{\pm}5.3$ points. The IKDC grade was abnormal(C) or severely abnormal(D) in 25 cases preoperatively They improved 22(88%) of normal(A) or nearly normal(B) and 3(12%) of abnormal. Conclusion: Short term results of ACL reconstruction using Achilles allograft was acceptable. Achilles allograft can be a reasonable graft alternative to autograft for ACL reconstruction.

• Journal of Yeungnam Medical Science
• /
• v.5 no.1
• /
• pp.111-125
• /
• 1988

The authors studied obsession, using Derogatis'SCL-90, and lie tendency using MMPI, in the subjects of 2586 male and 1206 female college freshmen of Yeungnam University. The authors collected the reports of self-rating Obsession scale and Lie scale during the periods from January to February, 1988, and applied ANOVA and t-test on obsession and lie scores in order to compare than between various psychosocial factors, and sexes. The results as follows: 1. There was not significant difference in the mean average of total obsession scores between male and female students: male students scored $10.26{\pm}5.21$, female students scored $10.54{\pm}5.01$. 2. High scored items were doubting, indecisiveness, inattention, obsessive thoughts and preoccupation with accuracy. 3. 37 male students(1.4%) showed high obsession scores of 25 or higher, while 9 female students(0.7%) showed thd same scores. 4. There was strong tendency toward higher obsession scores in the students who were dissatisfied with their home atmosphere, college, department and familiarity of parents and those who had pessimistic view of self in the past, present or future in both group(respectively P<0.001). The male students who attended the department of fine arts, who grew at the Up & whose educational fees were paid by brothers & sisters showed higher level of obsession scores(respectively P<0.05, P<0.05, P<0.01). The female students who grew at the Myun & who were self-cook showed higher level of obsession scores(respectively P<0.05). 5. There was not significant difference in the mean averages of total lie scores between male and female students: male students scored $4.73{\pm}2.49$, female students scored $4.70{\pm}2.16$. 6. High scored items were NO. 14, 8, 11, 9 and 1 in the table 6. 7. 98 male students(3.7%) showed high lie scores of 70 or high, while 27 female students(2.2%) showed the same scores. 8. There was strong tendency toward higher lie scores in the students who were satisfied with their home atmosphere, college, department and familiarity of parents and those who had optimistic view of self in the past, present or future in both group(respectively P<0.001). The students who attended the department of pharmacy, music, evening programs showed higher level of lie scores(P<0.05).

• Journal of Yeungnam Medical Science
• /
• v.9 no.2
• /
• pp.275-287
• /
• 1992

In order to survey the reality of anxiety and depression among the Koreans residing in China, a study was conducted between January and March of 1991, on the residents of Yun-Kil city, with subjects of 472 Koreans and 479 Chinese. The evaluation was based on the questionairs, named Combined self-rating anxiety depression scale(CADS), distributed among the subjects. ANOVA and t-test were applied for data processing. The results were as follows : There was not significant difference in the mean of total scores between the two groups. The scores of Koreans were $29.70{\pm}7.03$, while those of Chinese were $29.45{\pm}9.01$. The score of the CADS above 50(clinially significant level) was seen in 12(2.54%) Koreans and 21(4.38%) Chinese. The anxiety-depression scores relating to the items of indigestion and decreased appetite, sleep disturbance, apprehension, decreased libido were relatively high among the Koreans. The items appeared low in scores among the Koreans were faintness, fear, suicidal rumination, hopelessness, paresthesias. The highs among the Chinese were facial flushing, anxiousness, dissatisfaction, suicidal rumination. The items appeared low among the Chinese were fear, faintness, paresthesias, weight loss, suicidal rumination. In the comparison of evaluation by items between the two groups, the items placing the Koreans significantly higher over the Chinese are indigestion & decreased appetite, sleep disturbance, apprehension, decreased libido. The Chinese marked significantly higher in facial flushing, anxiousness, dissatisfaction, suicidal rumination. Those in the case of female (p<0.01 respectively), less than twenty years old (p<0.01 respectively), dissatisfied with family relationship(p<0.01 respectively), with past history of psychiatric hospitalization(Koreans p<0.01, Chinese p<0.05), pessimistic toward future, present, past self image(p<0.01 respectively) had significantly higher scores in both groups. In religion, neither group showed significant difference. In religion, neither group showed significant difference. In marital status, the Koreans showed a higher degree of divorce and separation and the Chinese in singleness(p<0.01 respectively). The Korean were higher in illiteracy and the Chinese had more college education(p<0.01 respectively). In place of growth, the Koreans showed not much difference in the areas while more Chinese grew up un large cities(p<0.01). More Koreans lived in the dormitory while the Chinese were engaged more in self-cooking(p<0.01 respectively). In pocket money per mouth, more Koreans were less than 1 dollar while the Chinese were between 7 and 10 dollars(p<0.01 respectively). There were no significant difference between two groups about religion.

• Korean Journal of Psychosomatic Medicine
• /
• v.14 no.1
• /
• pp.53-61
• /
• 2006

Objectives: This study was designed to evaluate difference of the alexithymia between panic patients and normal controls by examination of the relationships between different components of the alexithymia construct and level of anxiety and depression in panic patients and normal controls. Methods The subjects were 167 patients who met DSM-IV criteria for panic disorder and 110 normal controls. They drew up symptom checklists and self-rating scales, and were measured by Anxiety Disorders Interview Schedule-Panic attack & Agoraphobia(ADIS-P & A), Korean version of Toronto Alexithymia Scale (TAS-20K), Spielberger State-Trait Anxiety Inventory-State & Trait (STAI-S & T), Beck Depression Inventory (BDI), and Revised Anxiety Sensitivity Index (ASI-R). For statistical analysis, we performed t-test to compare the sociodemographic characteristics and the scores of self reported scales between panic patients and normal controls. Pearson correlation was performed between TAS-20K and it's subfactors, STAI-S & T, ASI-R and BDI in panic patients and normal controls. And stepwise multiple regression analysis was preformed to explain results of correlation analysis for alexithymia. Results: The panic patients reported more significant alexithymic (p<0.001), more difficulty identifying feeling (p<0.001) and describing feeling (p=0.001) than normal controls. Futhermore, panic patients were more significant anxious, sensitive to anxious feeling and depressive than normal controls. Moreover, the alexithymia of panic patients was explained by trait-anxiety $({\Delta}R^2=0.255)$ and anxiety sensitivity $({\Delta}R^2=0.062)$, that of normal controls was predicted by depression $({\Delta}R^2=0.144)$ and anxiety sensitivity $({\Delta}R^2=0.033)$ Conclusion: The panic patients reported more anxious and sensitive to anxious feeling, and these symptoms predict alexithymia in panic patients. However, the alexithymia of normal controls was explained by depression more than anxiety sensitivity, and such a result isn't consistent with previous studies and this may be mainly due to difference of the homogeneity in object of the studies.

• Korean Journal of Psychosomatic Medicine
• /
• v.7 no.2
• /
• pp.184-195
• /
• 1999

Objectives : As traumatic brain injury(TBI) leaves chronic sequelae in mind and body, the injured patients should rectify the meaning and object that they have pursued in their lives and set up a new purpose in life that they may make the rest of their lives meaningful. This study was designed to investigate the purpose and quality of life levels and the influence of demographic and clinical variables on the levels in the patients with TBI, and to be of some help to their rehabilitation. Methods : In order to assess the purpose in life(PIL) and the quality of life(QOL) levels, Purpose-in-Life Test, Sickness Impact Profile, Quality of Life Index, Head Injury Symptom Ckecklist, and Neurobehavioral Rating Scale were administered to the subjects. The subjects were thirty-two patients with TBI and the same numbered normal controls. The TBI group was composed of 16 to 65 year-aged patients who had received mild or severe TBI at least 12 months before, and the controls were siblings or friends of the patients whose age, sex, and educational level were similar to them. Results : 1) The PIL and QOL levels of the patients with TBI remained significantly lower than that of control group after their symptoms of injury were stabilized(p<.01, p<.01). 2) The mean PIL score of TBI group was $58.8{\pm}23.2$, which was to be regarded as the level of existential vacuum. 3) The PIL level of TBI group was significantly correlated with the QOL level(p <.01). 4) The subgroup with lower PIL level in patients with TBI has significantly higher rate of female than that with higher PIL(p<.05), the PIL level of female patients was significantly lower than that of male patients(p <.05). 5) The significant differences in PIL levels were not found, in which comparison was performed between each pair of subgroups of patients with TBI divided by severity of injury(mild vs severe), marital status(married vs unmarried), and occupational status prior to injury(employed vs unemployed). Conclusion : The PIL of patients with TBI still remained the level of existential vacuum after symptoms of sequelae had been stabilized, The QOL level was also extremely low, and as the PIL level was low the QOL was also low. The demographic and clinical variables except sex did not have influence on the PIL level in brain-injured patients. It is suggested that every patient should admit their mental and physical limitations caused by brain injury and revise their purpose in life for successful rehabilitation.

• Journal of The Korean Association For Science Education
• /
• v.22 no.4
• /
• pp.872-881
• /
• 2002

Student questioning is included in the priority of science literacy, to enable students to solve problems by exploring questions, communicating and constructing knowledge(AAAS, 1989). Also, the essence of student questioning in science lies in its function as a link between thinking and learning. But educators did not pay much attention to students' questioning in Korea. The purpose of this study was to investigate the effects of science question enhancement instruction on students' science questioning level and achievement. Also, this study showed the effects of other variables(logical thinking, science achievement, interest, and gender) on students' science questioning level. The pretest-posttest control group design group design was used. The sample was consisted of 80 second grade middle school students in experimental group(Science question enhancement instruction) and 74 students in control group(Traditional learning). Students in both groups were received identical content instruction on the unit 'Structures and functions of plant'. These groups were treated for 15 hours during 6 weeks. Students' questions were rated using the four levels described by the Middle School Students' Science Question Rating Scale(r= .96,)(Cuccio-Schirripa & Steinner, 2000). Science achievement data were collected using a 17 item multiple choice test(Cronbach ${\alpha}$= .84). To investigate students' logical thinking ability, a abridged GALT(Cronbach ${\alpha}$= .69) was used. Five-way ANOVA, ANCOVA, and multiple regression analysis were used to analyze the results. The results indicated that students who received instruction on researchable questioning outperformed those students who were not instructed on high-order questioning(p< .01). Results of correlations indicated that instruction(r= .640), science achievement(r= .311) and logical thinking ability(r= .212) was significantly and positively related with students' questioning level. But, interest and gender did not show any significant correlation with students' questioning level. Science question enhancement instruction was more effective on science achievement than the traditional instruction(p< .01).

• Journal of Hospice and Palliative Care
• /
• v.8 no.2
• /
• pp.216-223
• /
• 2005

Purpose: The purpose of this study was to compare the need for hospice care between families of children and adults with cancer. Methods: The data was collected from 190 families of children and adult with cancer using self-rating questionnaires from December, 2004 to February, 2005. Data was analyzed using SPSS/Win program by Mean, SD and t-test. Results: The mean score of the need for hospice care in families of children with cancer was greater significantly than in families of adults with cancer (t=-2.126, P=.035). The scores of two factors among the five factors evaluated for the need for hospice care were different significantly. The mean score of control of major terminal physical symptoms' in families of children with cancer was greater significantly than the mean score of adults with cancer (t=-2.165, P=.032). The mean score of 'spiritual care to prepare for death' in families of adults with cancer was greater significantly than the mean score of children with cancer (t=-2.380, P=.018). Conclusion: For improving the quality of life for both patients and families, the hospice service program needs to consider the life cycle of patients.

• Journal of Preventive Medicine and Public Health
• /
• v.28 no.1 s.49
• /
• pp.103-121
• /
• 1995

An investigation on psychophysiological health and social well-being of shift workers been carried out on workers of a automobile manufacturing plant in Ulsan, for 1 month from April, 1993. This cross-sectional survey compared shift workers(n=544) with day workers(n=115). Each subject completed a questionnaire about his personal habit, background, shift schedule, sleep and eating patterns, subjective digestive symptom and psychological well-being and distress using the General Well-Being Schedule(GWB) by self administrated questionnaire that was developed for the U.S. Health and Nutrition Examination Surveys (HANES I). Chi-square analysis was used for discontinuous data and the t-test was used for continous data to determine whether differences noted between the two groups. In terms of sleep quality, a greater percentage of shift workers frequently had trouble getting back to sleep once awakend(p<0.01), and a greater percentage of shift workers awakend tired or sleepy frequently more often than day workers(p<0.05). In rating the quality of their sleep, more shift workers rated their sleep fair to poor than day workers, and greater percentage of shift workers felt tired or sleepy at work two or more times per week(p<0.01) and a much higher percentage of shift workers felt tired or sleepy after work every days(p<0.01). In terms of sleep patterns, a much higher percentage of day workers reported uninterrupted sleep per 24hours than shift workers. The shift workers reported different eating patterns from day workers but there were no statistically significant and rate of their appetite. Thirteen percent of day workers reported the best appetite but only 6.6% of shift workers had the best appetite. The gastric complaints is more frequent in shift workers than day workers(p<0.01). Among subscores in General Well-Being Schedule, anxiety, depression, positive well-being and vitality subscale of shift workers were lower than those of day workers(p<0.05) and general health and self control subscale of shift workers were lower than .those of day workers but there were no statistical significant difference. Based on these study result, it could be concluded that the shift work has significant effects on some psychophysiological conditions of the workers.