• Journal of Intelligence and Information Systems
• /
• v.26 no.1
• /
• pp.23-45
• /
• 2020

Big data is creating in a wide variety of fields such as medical care, manufacturing, logistics, sales site, SNS, and the dataset characteristics are also diverse. In order to secure the competitiveness of companies, it is necessary to improve decision-making capacity using a classification algorithm. However, most of them do not have sufficient knowledge on what kind of classification algorithm is appropriate for a specific problem area. In other words, determining which classification algorithm is appropriate depending on the characteristics of the dataset was has been a task that required expertise and effort. This is because the relationship between the characteristics of datasets (called meta-features) and the performance of classification algorithms has not been fully understood. Moreover, there has been little research on meta-features reflecting the characteristics of multi-class. Therefore, the purpose of this study is to empirically analyze whether meta-features of multi-class datasets have a significant effect on the performance of classification algorithms. In this study, meta-features of multi-class datasets were identified into two factors, (the data structure and the data complexity,) and seven representative meta-features were selected. Among those, we included the Herfindahl-Hirschman Index (HHI), originally a market concentration measurement index, in the meta-features to replace IR(Imbalanced Ratio). Also, we developed a new index called Reverse ReLU Silhouette Score into the meta-feature set. Among the UCI Machine Learning Repository data, six representative datasets (Balance Scale, PageBlocks, Car Evaluation, User Knowledge-Modeling, Wine Quality(red), Contraceptive Method Choice) were selected. The class of each dataset was classified by using the classification algorithms (KNN, Logistic Regression, Nave Bayes, Random Forest, and SVM) selected in the study. For each dataset, we applied 10-fold cross validation method. 10% to 100% oversampling method is applied for each fold and meta-features of the dataset is measured. The meta-features selected are HHI, Number of Classes, Number of Features, Entropy, Reverse ReLU Silhouette Score, Nonlinearity of Linear Classifier, Hub Score. F1-score was selected as the dependent variable. As a result, the results of this study showed that the six meta-features including Reverse ReLU Silhouette Score and HHI proposed in this study have a significant effect on the classification performance. (1) The meta-features HHI proposed in this study was significant in the classification performance. (2) The number of variables has a significant effect on the classification performance, unlike the number of classes, but it has a positive effect. (3) The number of classes has a negative effect on the performance of classification. (4) Entropy has a significant effect on the performance of classification. (5) The Reverse ReLU Silhouette Score also significantly affects the classification performance at a significant level of 0.01. (6) The nonlinearity of linear classifiers has a significant negative effect on classification performance. In addition, the results of the analysis by the classification algorithms were also consistent. In the regression analysis by classification algorithm, Naïve Bayes algorithm does not have a significant effect on the number of variables unlike other classification algorithms. This study has two theoretical contributions: (1) two new meta-features (HHI, Reverse ReLU Silhouette score) was proved to be significant. (2) The effects of data characteristics on the performance of classification were investigated using meta-features. The practical contribution points (1) can be utilized in the development of classification algorithm recommendation system according to the characteristics of datasets. (2) Many data scientists are often testing by adjusting the parameters of the algorithm to find the optimal algorithm for the situation because the characteristics of the data are different. In this process, excessive waste of resources occurs due to hardware, cost, time, and manpower. This study is expected to be useful for machine learning, data mining researchers, practitioners, and machine learning-based system developers. The composition of this study consists of introduction, related research, research model, experiment, conclusion and discussion.

• Food Science and Preservation
• /
• v.23 no.3
• /
• pp.402-412
• /
• 2016

Generation of biogenic amines (BAs) and physiological activities of Makgeolli, Korean rice wine, were investigated during fermentation under different temperatures (20, 25, and $30^{\circ}C$) and time (3, 5, and 7 days). The pH was 3.96~4.36 during fermentation and the acidity increased proportionally with temperature and time. Alcohol and total phenolic contents peaked on day 5 at 25 by 9.0~9.8% and $3.01{\pm}0.07mg/g$, respectively. The organic acid mostly produced was lactic acid, which increased gradually with temperature and time. Formation of biogenic amines comprised of tryptamine and putrescine was observed during fermentation: 6.91, 11.70, and 15.63 mg/mL of tyramine on day 7 at 20, 25, and $30^{\circ}C$, respectively. Antioxidant activities from DPPH and FRAP assay were high in the order of $25^{\circ}C$>$20^{\circ}C$>$30^{\circ}C$ while that from ABTS assay was high regardless of temperature. Processed at 1 mg/mL and $30^{\circ}C$, the inhibitory effect on ${\alpha}$-amylase was $67.99{\pm}0.11$, $73.64{\pm}0.43$, and $75.51{\pm}0.26%$ on days 3, 5, and 7, respectively, which increased proportionally with temperature and time. Inhibitory activity on lipid accumulation in 3T3-L1 adipocytes was the highest in Makgeolli fermented on day 5 at $25^{\circ}C$. This study shows that fermentation at $25^{\circ}C$ for 5 days yields Makgeolli of the best quality with high bioactivities but no biogenic amines.

• Journal of the Korean Society of Food Science and Nutrition
• /
• v.41 no.1
• /
• pp.79-86
• /
• 2012

This study was performed to investigate the physiochemical properties and anti-diabetic effect of yacon vinegar by two-step fermentation. Yacon was matured at room temperature for 20 days. The sugar content of yacon juice prepared from mature yacon was approximately $14^{\circ}Brix$. In the first stage, yacon wine was produced from the juice at $28^{\circ}C$ for 6 days. In the second stage, acetic acid fermentation was conducted at $30^{\circ}C$ and 200 rpm for 6 days to produce yacon vinegar with 4.75% acidity. The major free sugars of yacon vinegar were glucose and fructose at 2,072.12 mg% and 463.95 mg%, respectively. The acetic acid content was the highest of the major organic acids at 3,881.44 mg%. The total free amino acid content was 62.88 mg% with the main free amino acids being proline, ${\gamma}$-amino-n-butyric acid and ornithine. The major minerals of yacon vinegar were Ca, K and Mg. The in vivo anti-diabetic activity of yacon vinegar was investigated in high-fat diet (HFD)/streptozotocin (STZ)-induced diabetic mice. Diabetic mice were administered orally with 10% yacon juice and two yacon vinegars (5% and 10%) at a dose of 7 mL/kg body weight once per day for 4 weeks. Five% yacon vinegar improved the fasting blood glucose levels and glucose tolerance test significantly compared to the diabetic control group (p<0.05). Yacon vinegar increased the pancreatic C-peptide concentration in a dose-dependent manner. These results show that 5% yacon vinegar has a more potent effect on ameliorating hyperglycemia than 10% yacon juice.

• Journal of the Korean Society of Food Science and Nutrition
• /
• v.41 no.1
• /
• pp.136-142
• /
• 2012

Sogokju, a Korean glutinous rice wine and one of the oldest Korean traditional wines, is famous for its unique taste acquired from a 100-day incomplete fermentation process. This study investigated the effects of the degree of rice milling on the physicochemical and sensory characteristics of Sogokju. It evaluated the physicochemical characteristics, pasting and color properties, and structural properties of starch using four different degrees of milled rice (Oryza sativa L.) cultivar Dongjinchalbyeo. Samples of brown rice with milling yields of 92%, 84%, 76%, and 68% were produced using both abrasive and friction whiteners. This study showed that the protein, lipid, and ash content of milled rice decreased as the degree of milling increased. The lower hardness of the kernel below milling yield 92% suggested that milling may be related to the lower protein content of the kernel. The pasting curve showed a significant increase in viscosity properties as the degree of milling increased. This is due to the decrease in protein and lipid content, the increase in starch content, and the difference in amylopectin chain-length distribution. Further milling of white rice, based on 92% milling yield, had an effect on the amylopectin chain-length distribution due to the degree of polymerization (DPn) of 37~60. The long chain of amylopectin also contributed to the viscosity. The increase in the degree of milling decreased the glucose and total sugar content of Sogokju. However, it increased the total acidity of Sogokju. Moreover, the lightness of Sogokju decreased while its yellowness increased. These results indicate that the degree of milling can alter the taste and color of Sogokju. The sensory evaluation showed that the increase in the degree of milling decreased consumer preference for Sogokju. The sensory score for Sogokju was positively correlated with its brix degree, glucose content, pH, and protein content of raw rice.

• Journal of Fisheries and Marine Sciences Education
• /
• v.23 no.3
• /
• pp.410-424
• /
• 2011

This study was investigated for the purpose of obtaining basic data which can be applied to processing of canned sea mussel using tomato paste. Shell were washed, and then steamed and shucked. Sea mussel meat was prepared with ratio of sea mussel 90g, tomato paste sauce 65g(tomato paste 42%, gum guar 1.0%, salt 2.0%, starch syrup 2.0%, cooking wine 1%, water 52%). The sea mussel meats were packed with vacuum seamer in 301-3 can, and then sterilized for various F0 value(F0 8-12 min.) in a steam system retort at $118^{\circ}C$. The factors such as pH, VBN, amino-N, total amino acid, free amino acid, chemical composition, color value (L, a, b), texture profile, TBA value, mineral, sensory evaluation and viable bacterial count of the canned sea mussel produced with various sterilization condition(F0 8-12 min.) were measured. The same element was also measured during preservation. The results showed that the product sterilized at F0 8 min. and preserved for 90 days were the most desirable.

• Food Science and Preservation
• /
• v.23 no.6
• /
• pp.797-803
• /
• 2016

In this study, we investigated the effect of fermentation conditions on the amylolytic and proteolytic activities of Aspergillus luchuensis strain 74-5 and Aspergillus oryzae strain 75-2, which are used in the preparation of the starter culture, for Takju (Korean traditional rice wine). The starter culture was optimized using different conditions, such as inoculum size, inoculation temperature, and incubation time. The enzyme activities under each condition were measured. In the A. luchuensis strain 74-5 starter culture, the ${\alpha}-amylase$ and glucoamylase activities increased, however the activity of acidic protease decreased as the diluent to starter culture ratio increased. In the A. oryzae 75-2 starter culture, all enzyme activities were maintained at a higher level even at 5% inoculation ratio. Higher enzyme activities were observed in the middle range of inoculation temperature (35, $40^{\circ}C$), than in the lower range (20, $30^{\circ}C$). Enzyme activity in the starter culture varied with incubation time, however it was the highest at 144 and 120 hr, respectively, for A. luchuensis strain 74-5 and A. oryzae strain 75-2. The spore count of the starter culture was approximately $2{\times}10^7$ during fermentation, out of which contamination by aerobic bacteria was about $3{\times}10^3$. The results suggested that the starter culture of each strain could be used as an inoculum for fermentation. However, we needs to conduct further research for the selection of suitable diluting agents as well as drying methods to reduce the contamination by aerobic bacteria, while retaining the enzyme activity.

• Microbiology and Biotechnology Letters
• /
• v.35 no.3
• /
• pp.238-244
• /
• 2007

To identify the useful biological activities and the superiority in quality of Korean traditional liquors, we prepared the alcohol-free concentrates from different types of commercial traditional liquors, Takju ($T1{\sim}T3$), Yakju/Cheongju ($Y-1{\sim}Y-5$), Fruit wine (F-1) and Soju (S-1, S-2), respectively. We investigate their compositions as well as antimicrobial, antioxidant, and anti-thrombosis activity. Among the 11 traditional liquors, Y-3, Y-4, Y-5 and F-1 showed higher total-polyphenol, total-flavonoids and reducing sugars than the others. The strong antibacterial and anti-thrombosis activities were identified in Y-3, Y-4, Y-5 and F-1, and a minor antioxidant activity was found in F-l. The antibacterial activity of the Y-3, Y-4, Y-5 and F-1 alcohol-free concentrates showed a broad-spectrum, and growth inhibition was found in gram-positive, gram-negative, and ampicillin-resistant bacteria. The sequential solvent fractionation of Y-3, Y-4, Y-5 and F-1, and following analysis showed that ethyl acetate fractions of Y-3, Y-4, Y-5 and F-1 possess strong antibacterial and anti-thrombosis activity. Especially, the ethyl acetate fractions of Y-3, Y-4 and F-1 showed superior anti-thrombosis activity compared than that of aspirin. Our results suggest that the useful substances are produced from substrates and edible plant added during the fermentation, and the Korean traditional liquors could be developed as strong antibacterial and anti-thrombosis agents.