• Science of Emotion and Sensibility
• /
• v.16 no.1
• /
• pp.53-64
• /
• 2013

Folksonomies have potential problems caused by synonyms, tagging level, neologisms and so forth when retrieving music by tags. These problems can be tackled by introducing the mood intensity (Arousal and Valence value) of music as its internal tag. That is, if moods of music pieces and their mood tags are all represented internally by numeric values, A (Arousal) value and V (Valence) value, and they are retrieved by these values, then music pieces having similar mood with the mood tag of a query can be retrieved based on the similarity of their AV values though their tags are not exactly matched with the query. As a prerequisite study, in this paper, we propose the mapping table defining the relation between AV values and folksonomy tags. For analysis of the association between AV values and tags, ANOVA tests are performed on the test data collected from the well known music retrieval site last.fm. The results show that the P values for A values and V values are 0.0, which means the null hypotheses could be rejected and the alternative hypotheses could be adopted. Consequently, it is verified that the distribution of AV values depends on folksonomy tags.

• Korean journal of food and cookery science
• /
• v.21 no.1 s.85
• /
• pp.64-74
• /
• 2005

To develop the diet for Korean type 2 diabetic patients, a consumer survey was conducted (N=35) and was analyzed statistically. The survey compared the results of a questionnaire, somatometry, and blood parameters before and 8 weeks after the experiment. It was divided into Diabetic diet enforcement group (DDE) vs. Non diabetic diet enforcement group (NDDE). There was no significant difference in body weight, body mass index (BMI), body fat, triglyceride (TG) and low density lipoprotein cholesterol (LDL-C) of the subjects before and 8 weeks after the experiment. DDE's average fasting glucose (FBG) and post prandial 2 hours blood glucose (PP2) were significantly decreased before and after the experiment (p<0.05). NDDE's average FBG and PP2 showed no significant difference. In terms of Hemoglobin $A_lC\;(HbA_1C)$ content, DDE's had significantly decreased level both before and after the experiment (p<0.05), while there was no significant difference in the NDDE group. Cholesterol and High density lipoprotein cholesterol (HDL-C) were significantly different in DDE (p<0.05) both before and after the experiment. Whereas in the case of NDDE, there was no significant difference on the cholesterol and HDL-C, while DDE's showed a significant difference before and after the experiment. In terms of patients distribution depending on their treatment (p<0.05), NDDE had no significant difference before and after the experiment. For an item asking blood glucose control, DDE had significant difference before and after the experiment (p<0.01), while NDDE had no significant difference. For program satisfaction (p<0.05) and health (p<0.001), both the groups changed significantly after the experiment. But there was no significant difference in applying it to their real life after the experiment. Based on these results, DDE had the decreased blood glucose levels, $HbA_1C$, and cholesterol and increased HDL-C, with decreased rate of the patients taking oral hypoglycemic agents. In addition, many of the patients who participated in the experiment were found to be satisfied with the program, in terms of having less trouble in glucose control and exhibited improvement in health. Hence, based on the above results it was concluded that program was a very successful one for the treatment of diabetes.