• Journal of Intelligence and Information Systems
• /
• v.27 no.1
• /
• pp.23-46
• /
• 2021

Collaborative filtering, which is often used in personalization recommendations, is recognized as a very useful technique to find similar customers and recommend products to them based on their purchase history. However, the traditional collaborative filtering technique has raised the question of having difficulty calculating the similarity for new customers or products due to the method of calculating similaritiesbased on direct connections and common features among customers. For this reason, a hybrid technique was designed to use content-based filtering techniques together. On the one hand, efforts have been made to solve these problems by applying the structural characteristics of social networks. This applies a method of indirectly calculating similarities through their similar customers placed between them. This means creating a customer's network based on purchasing data and calculating the similarity between the two based on the features of the network that indirectly connects the two customers within this network. Such similarity can be used as a measure to predict whether the target customer accepts recommendations. The centrality metrics of networks can be utilized for the calculation of these similarities. Different centrality metrics have important implications in that they may have different effects on recommended performance. In this study, furthermore, the effect of these centrality metrics on the performance of recommendation may vary depending on recommender algorithms. In addition, recommendation techniques using network analysis can be expected to contribute to increasing recommendation performance even if they apply not only to new customers or products but also to entire customers or products. By considering a customer's purchase of an item as a link generated between the customer and the item on the network, the prediction of user acceptance of recommendation is solved as a prediction of whether a new link will be created between them. As the classification models fit the purpose of solving the binary problem of whether the link is engaged or not, decision tree, k-nearest neighbors (KNN), logistic regression, artificial neural network, and support vector machine (SVM) are selected in the research. The data for performance evaluation used order data collected from an online shopping mall over four years and two months. Among them, the previous three years and eight months constitute social networks composed of and the experiment was conducted by organizing the data collected into the social network. The next four months' records were used to train and evaluate recommender models. Experiments with the centrality metrics applied to each model show that the recommendation acceptance rates of the centrality metrics are different for each algorithm at a meaningful level. In this work, we analyzed only four commonly used centrality metrics: degree centrality, betweenness centrality, closeness centrality, and eigenvector centrality. Eigenvector centrality records the lowest performance in all models except support vector machines. Closeness centrality and betweenness centrality show similar performance across all models. Degree centrality ranking moderate across overall models while betweenness centrality always ranking higher than degree centrality. Finally, closeness centrality is characterized by distinct differences in performance according to the model. It ranks first in logistic regression, artificial neural network, and decision tree withnumerically high performance. However, it only records very low rankings in support vector machine and K-neighborhood with low-performance levels. As the experiment results reveal, in a classification model, network centrality metrics over a subnetwork that connects the two nodes can effectively predict the connectivity between two nodes in a social network. Furthermore, each metric has a different performance depending on the classification model type. This result implies that choosing appropriate metrics for each algorithm can lead to achieving higher recommendation performance. In general, betweenness centrality can guarantee a high level of performance in any model. It would be possible to consider the introduction of proximity centrality to obtain higher performance for certain models.

• Clinical and Experimental Pediatrics
• /
• v.48 no.10
• /
• pp.1055-1060
• /
• 2005

Purpose : The purpose of this study is to compare the growth pattern of breast fed and formula fed infants in the first 1 year of life. Methods : Anthropometric data(weight, length, head circumference) of at birth, 1, 3, 6, 9 and 12 months were collected by chart review and characteristics of subjects were collected by questionnaires. Among 358 infants, breast fed infants were 161(84 males, 77 females) and formula fed infants were 90(42 males, 48 females). Neither group was given solid foods before 4 months. The weight for age, length for age and head circumference for age were calculated. Breast fed infants were separated into 2 groups(breast fed for 4-11 months and breast fed for more than 12 months). Results : Characteristics of infants and mothers were similar in both groups except for maternal age. Mean weight of breast fed group was lower than that of formula fed group at 12 months of age(male : P=0.004, female : P=0.004). However, mean weight of 12 months breast fed group was below formula fed groups weight at 9 and 12 months(P<0.05). Mean length and head circumference were similar between groups. Conclusion : The growth indices of breast fed and formula fed infants are similar at birth, but weight curves of two groups differ in the first 1 year.

• Korean Journal of Food Science and Technology
• /
• v.34 no.4
• /
• pp.552-558
• /
• 2002

Polyphenol oxidase from Flammulina velutipes was purified and characterized. Purification of polyphenol oxidase was achieved by ammonium sulfate precipitation, Superdex G-200 gel filtration chromatography, Phenyl superose affinity chromatography, Mono-Q anion exchange chromatography and Superdex S-200 gel filtration chromatography on FPLC. After these purification steps specific activity of purified polyphenol oxidase increased to 199.1 units/mg. Polyphenol oxidase from F. velutipes was composed of a single polypeptide with molecular weight of about 40 kDa. Optimum pH and temperature for the enzyme reaction were found to be 6.0 and $25^{\circ}C$, respectively. The activity of the enzyme gradually decreased at acidic pH between 3 and 5, and the enzyme lost its activity at alkaline pH between 8 and 10. This enzyme exhibited high substrate specificity to o-diphenols. Km-values for L-DOPA and caffeic acid were found to be 3.97 mM and 1.78 mM, respectively. 2-mercaptoethanol, L-ascorbic acid, sodium bisulfite, EDTA and $Mg^{2+}$ inhibited the activity of pholyphenol oxidase and $Cu^{2+}$, $Fe^{2+}$, $Zn^{2+}$ and $Ni^{2+}$ increased enzyme activity. The activity of enzyme was well maintained at $-70^{\circ}C$ for over 4 months, and at $-20^{\circ}C$ for 1 months.

• Journal of Korean Society of Forest Science
• /
• v.87 no.3
• /
• pp.391-403
• /
• 1998

The study was carried out to devise a proper measure to increase the income of mountain villagers by producing sap water of Acer mono, and to make the most of sap water as local specialty to contribute to the local economy of mountain villages. All the processes from collecting to marketing of sap water of Acer mono was investigated. The survey was done from mid-January to mid-February in the 3 major sap water collecting regions, Toji-myon Kurey-gun(Piagol area of Mt. Chiri), Okryong-myon Kwangyang city(Mt. Baekun), and Jookhack-ri Sunchon(Mt. Chokey). A total of 90 householders who collect sap water, to say again, 30 householders in each region, were interviewed personally to make up questionnaires. The habitual or general practices about collecting sap water, the selling price, the sales process, labor power to collect and carry down, carrying distance and facilities, sales income and side income, and family income were investigated and examined. Spots of collecting sap water were not concentrated but scattered all over the collecting area. Collecting method, collecting amount, sales process, and selling price varied with the village and region. Sap water was collected by tapping or boring method, the latter of which was widely used in lots of regions except in Sunchon. Although the amount of sap production per family varied with region, the average amount was about 1,350 liters. Of all the sap water collected, 44% was consumed by drinking of on-the-spot visitors and 36% was sold by order, etc. Sap water was sold at the price varying from 10,000 won to 60,000 won per 18 liters. The average selling price was 41,000 won, but selling prices of 43,000 won and 45,000 wan amounted to 38% and 25%, respectively.