• Journal of Intelligence and Information Systems
• /
• v.24 no.2
• /
• pp.85-109
• /
• 2018

Recommender system recommends the items expected to be purchased by a customer in the future according to his or her previous purchase behaviors. It has been served as a tool for realizing one-to-one personalization for an e-commerce service company. Traditional recommender systems, especially the recommender systems based on collaborative filtering (CF), which is the most popular recommendation algorithm in both academy and industry, are designed to generate the items list for recommendation by using 'overall rating' - a single criterion. However, it has critical limitations in understanding the customers' preferences in detail. Recently, to mitigate these limitations, some leading e-commerce companies have begun to get feedback from their customers in a form of 'multicritera ratings'. Multicriteria ratings enable the companies to understand their customers' preferences from the multidimensional viewpoints. Moreover, it is easy to handle and analyze the multidimensional ratings because they are quantitative. But, the recommendation using multicritera ratings also has limitation that it may omit detail information on a user's preference because it only considers three-to-five predetermined criteria in most cases. Under this background, this study proposes a novel hybrid recommendation system, which selectively uses the results from 'traditional CF' and 'CF using multicriteria ratings'. Our proposed system is based on the premise that some people have holistic preference scheme, whereas others have composite preference scheme. Thus, our system is designed to use traditional CF using overall rating for the users with holistic preference, and to use CF using multicriteria ratings for the users with composite preference. To validate the usefulness of the proposed system, we applied it to a real-world dataset regarding the recommendation for POI (point-of-interests). Providing personalized POI recommendation is getting more attentions as the popularity of the location-based services such as Yelp and Foursquare increases. The dataset was collected from university students via a Web-based online survey system. Using the survey system, we collected the overall ratings as well as the ratings for each criterion for 48 POIs that are located near K university in Seoul, South Korea. The criteria include 'food or taste', 'price' and 'service or mood'. As a result, we obtain 2,878 valid ratings from 112 users. Among 48 items, 38 items (80%) are used as training dataset, and the remaining 10 items (20%) are used as validation dataset. To examine the effectiveness of the proposed system (i.e. hybrid selective model), we compared its performance to the performances of two comparison models - the traditional CF and the CF with multicriteria ratings. The performances of recommender systems were evaluated by using two metrics - average MAE(mean absolute error) and precision-in-top-N. Precision-in-top-N represents the percentage of truly high overall ratings among those that the model predicted would be the N most relevant items for each user. The experimental system was developed using Microsoft Visual Basic for Applications (VBA). The experimental results showed that our proposed system (avg. MAE = 0.584) outperformed traditional CF (avg. MAE = 0.591) as well as multicriteria CF (avg. AVE = 0.608). We also found that multicriteria CF showed worse performance compared to traditional CF in our data set, which is contradictory to the results in the most previous studies. This result supports the premise of our study that people have two different types of preference schemes - holistic and composite. Besides MAE, the proposed system outperformed all the comparison models in precision-in-top-3, precision-in-top-5, and precision-in-top-7. The results from the paired samples t-test presented that our proposed system outperformed traditional CF with 10% statistical significance level, and multicriteria CF with 1% statistical significance level from the perspective of average MAE. The proposed system sheds light on how to understand and utilize user's preference schemes in recommender systems domain.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.17 no.8
• /
• pp.236-250
• /
• 2016

This study was conducted to provide baseline data regarding functional cosmetics so that Inula. based cosmetics can increase its competitiveness in the market as well as to understand current trends to enable anticipation of demands for future product development. For this research, general consumers over the age of 20 residing in Seoul and the Gyeonggi district were surveyed. The results show consumers preferred serum-type products among various types of cosmetics, and that they purchased these once every 1-3 months. Consumers also preferred these products in less than 10-30ml capacity, and at costs of less than 30,000-50,000 KRW. For whitening, functional cosmetics consumers also preferred the serum type, in less than 30-50ml capacity and priced less than 30,000-50,000 KRW. Consumers preferred to purchase functional cosmetics in single units. The major purchasing location, with a high preference rate, was cosmetic stores, and the major sources of information, also with high preference rates, were 'experienced reviews from family, friends and acquaintances' and 'TV advertisements'. Respondents selected 'over 50,000 KRW' the most for all items when responding to 'Purchase Intent for Functional Cosmetics containing Inula', and responded that they were willing to pay 10%-30% more for functional cosmetics containing Inula compared to standard functional cosmetics. These results show that businesses in the cosmetics industry need to take consumer demand into account when developing new functional cosmetic products, as well as establish plans to create specialized spaces that provide better quality service and increase word of mouth effect through better utilization of various types of offline media, social media, and blogs. The study also shows a need for businesses to develop products fully utilizing the Inula flower, which has been shown to be effective as a natural skin whitener, wrinkle reducer and skin moisturizer, to appeal to the increasing number of customers interested in health and beauty.

• Journal of Animal Science and Technology
• /
• v.46 no.4
• /
• pp.667-676
• /
• 2004

Yogurt was prepared from skim milk added with 1, 2, 4 or 6% of rice- or skim milk powders and commercial lactic acid bacterial starters. Changes in pH, titratable acidity, viable cell counts, viscosity, organic acid contents and carbohydrates during fermentation were monitored and its sensory evaluation was also performed. The optimum level of additives such as rice- and skim milk powders for yogurt manufacture was selected. Fermentation properties of yogurt added with rice and skim milk powders were studied. The control yogurt reached pH 4.5 after 10 hours of fermentation, whereas the samples added with 4 or 6% rice powders reached pH 4.5 in 6 hours and those added with skim milk powder reached in 8 hours. After 4 hours of fermentation, the control yogurt reached a titratable acidity at 1.0 %, whereas other samples exceeded 1.0%. After 4 hours of fermentation, the viable cell counts in the samples added with 4% rice powders were the highest. At the level of 4% rice powders, the curd viscosity decreased. Except for the sample added with 6% skim milk powders, all the samples produced higher lactic acid during fermentation, and galactose level in yogurt was the highest when added with 4% rice powder or 6% skim milk powder. In overall sensory evaluation using 5 parameters, the sensory scores of all the yogurts were not significantly different. However, the sample added with 4% skim milk powders was the best for color and overall acceptability. The sample added with 4% rice powders was the best for the flavor and texture. From this experiment, the optimun level of additives such as rice or skim milk powders was selected to be 4%.

• Journal of Animal Science and Technology
• /
• v.46 no.4
• /
• pp.677-686
• /
• 2004

The objective of this experiment was to select the best strain of lactic acid bacteria for the manufacture of new type of yogurt with rice powders. Changes in pH, titratable acidity, viable cell counts, viscosity, organic acid contents, carbohydrates during fennentation were monitored and sensory evaluation were examined. The yogurt added with 4% rice or skim milk powders and L. salivarius ssp. salivarius culture did not reach pH 4.5, because the production of acids in this media for the culture was weak. The yogurt added with 4% rice or skim milk powder with L. casei, the pH was low and the titratable acidity was high, and therefore the quality of yogurt after 8 hours from fermentation was not high. The yogurt added with 4% rice or skim milk powders with a mixed culture of B. longum, L. acidophilus, Streptococcus salivarius ssp. thermophilus was considered best for achieving pH 4.5 and titratable acidity of 1.0 % from 8 to 14 hours. The yogurt with a mixed culture had more acetic acid. Galactose was accumulated when L. salivarius ssp. salivarius or the mixed culture were used for fermenting yogurt. In sensory evaluation, the yogurt with the mixed culture received high overall sensory score. From these results, a mixed culture of B. longum, L. acidophilus, Streptococcus salivarius ssp. thermophilus was identified as the best for the manufacture of yogurt added with rice powder.

• Journal of the East Asian Society of Dietary Life
• /
• v.19 no.6
• /
• pp.967-974
• /
• 2009

The effects of Opuntia ficus-indica var. saboten powder on the growth of lactic acid bacteria (LAB) were investigated in order to explore the possibility of manufacturing fermented milk containing the powder. Differences in pH, acidity, LAB counts, viscosity, and sensory evaluation were measured. Also the effects of dietary supplementation on the growth of piglets were evaluated by feeding fermented milks containing 0.1%, 0.2%, 0.5%, and 1.0% Opuntia ficus-indica var. saboten powder. Levels of pH, titratable acidity, viable LAB counts and viscosity were significantly different by the addition of the powder. When fermented milks containg 0.1%, 0.2%, 0.5% and 1.0% Opuntia ficus-indica var. saboten powder were kept at $4^{\circ}C$ for 30 days, viable LAB counts remained high after 30 days of storage. The effects of dietary supplementation of 0.2% Opuntia ficus-indica var. saboten fermented milk were investigated by examining piglet growth rate and fecal ammonia gas release. The piglets were fed 100g/herd of the fermented milk for 14 days. Average daily body weight gain was significantly (p<0.05) improved (110%) with dietary supplementation of the fermented milk, compared to a control group. More over, fecal ammonia gas emissions were reduced by dietary supplementation of the Opuntia ficus-indica var. saboten fermented milk. Sensory evaluation results showed that the samples containing 0.1% and 0.2% Opuntia ficus-indica var. saboten powder had the most parameters similar to those of the control.

• Journal of Life Science
• /
• v.28 no.11
• /
• pp.1347-1353
• /
• 2018

The fermentation of non-digestible soy meal can convert polysaccharides into many compounds that have a wide variety of biological functions. Bacillus strains are capable of hydrolyzing non-digestible saccharides, such as melibiose, raffinose, and stachyose, found in soy meal components. A highly active ${\alpha}$-galactosidase (${\alpha}$-d-galactoside galactohydrolase, EC 3.2.1.22) was isolated from a bacterium in a traditional Korean fermented medicinal herb preparation. The isolate, T2-16, was identified as Bacillus coagulans based on its 16S rRNA sequence and biochemical properties, and the strain was named Bacillus coagulans NRR-1207. When incubated in 10%(w/v) skim milk, Bacillus coagulans NRR1207 caused a decrease in the pH of the culture medium, as well as an increase in titratable acidity and viable cell counts. This strain also showed higher activities of ${\alpha}$-galactosidase, ${\beta}$-galactosidase, ${\alpha}$-glucosidase, naphthol-AS-BO-phosphohydrolase, and acid phosphatase when compared to other enzymes. It hydrolyzed oligomeric substrates, such as raffinose and stachyose, and liberated galactose, indicating that the Bacillus coagulans NRR1207 ${\alpha}$-galactosidase hydrolyzed the ${\alpha}$-1,6 glycoside linkage. These results suggest that the decreased stachyose and raffinose contents observed in fermented soy meal are due to this ${\alpha}$-galactosidase activity. Bacillus coagulans NRR1207 therefore has potential probiotic activity and could be utilized in feed manufacturing, as well as for hydrolyzing non-digestible soy meal components.