• Journal of Intelligence and Information Systems
• /
• v.16 no.4
• /
• pp.159-172
• /
• 2010

The recommender system is one of the possible solutions to assist customers in finding the items they would like to purchase. To date, a variety of recommendation techniques have been developed. One of the most successful recommendation techniques is Collaborative Filtering (CF) that has been used in a number of different applications such as recommending Web pages, movies, music, articles and products. CF identifies customers whose tastes are similar to those of a given customer, and recommends items those customers have liked in the past. Numerous CF algorithms have been developed to increase the performance of recommender systems. Broadly, there are memory-based CF algorithms, model-based CF algorithms, and hybrid CF algorithms which combine CF with content-based techniques or other recommender systems. While many researchers have focused their efforts in improving CF performance, the theoretical justification of CF algorithms is lacking. That is, we do not know many things about how CF is done. Furthermore, the relative performances of CF algorithms are known to be domain and data dependent. It is very time-consuming and expensive to implement and launce a CF recommender system, and also the system unsuited for the given domain provides customers with poor quality recommendations that make them easily annoyed. Therefore, predicting the performances of CF algorithms in advance is practically important and needed. In this study, we propose an efficient approach to predict the performance of CF. Social Network Analysis (SNA) and Artificial Neural Network (ANN) are applied to develop our prediction model. CF can be modeled as a social network in which customers are nodes and purchase relationships between customers are links. SNA facilitates an exploration of the topological properties of the network structure that are implicit in data for CF recommendations. An ANN model is developed through an analysis of network topology, such as network density, inclusiveness, clustering coefficient, network centralization, and Krackhardt's efficiency. While network density, expressed as a proportion of the maximum possible number of links, captures the density of the whole network, the clustering coefficient captures the degree to which the overall network contains localized pockets of dense connectivity. Inclusiveness refers to the number of nodes which are included within the various connected parts of the social network. Centralization reflects the extent to which connections are concentrated in a small number of nodes rather than distributed equally among all nodes. Krackhardt's efficiency characterizes how dense the social network is beyond that barely needed to keep the social group even indirectly connected to one another. We use these social network measures as input variables of the ANN model. As an output variable, we use the recommendation accuracy measured by F1-measure. In order to evaluate the effectiveness of the ANN model, sales transaction data from H department store, one of the well-known department stores in Korea, was used. Total 396 experimental samples were gathered, and we used 40%, 40%, and 20% of them, for training, test, and validation, respectively. The 5-fold cross validation was also conducted to enhance the reliability of our experiments. The input variable measuring process consists of following three steps; analysis of customer similarities, construction of a social network, and analysis of social network patterns. We used Net Miner 3 and UCINET 6.0 for SNA, and Clementine 11.1 for ANN modeling. The experiments reported that the ANN model has 92.61% estimated accuracy and 0.0049 RMSE. Thus, we can know that our prediction model helps decide whether CF is useful for a given application with certain data characteristics.

• Journal of the Korean Magnetics Society
• /
• v.18 no.3
• /
• pp.109-114
• /
• 2008

The samples were synthesized by using a solid state reaction. The X-ray diffraction pattern for $Ti_{0.96}Co_{0.02}Fe_{0.02}O_2$ showed a pure rutile phase with tetragonal structure, Mixtures of the proper proportions of the elements sealed in evacuated quartz ampoule were heated at $870{\sim}930^{\circ}C$ for one day and then slowly cooled down to room temperature at a rate of $10^{\circ}C$/h. In order to obtain single phase material, it was necessary to grind the sample after the first firing and to press the powders into pellets before annealing them for a second time in evacuated and sealed quartz ampoule. Magnetic properties have been investigated using the vibrating sample magnetometer (VSM). Room temperature magnetic hysteresis (M-H) curve showed an obvious ferromagnetic behavior and the magnetic moment per Fe atom under the applied of 0.8 T was estimated to be about $1.5\;{\mu}_B$/CoFe. But the magnetic moment per Fe atom under the applied of 0.8 T was estimated to be about $0.02\;{\mu}_B$/CoFe without Ti-getter. Size of particles is about $1\;{\mu}m$ using the transmission electron microscope (TEM). The ingredients of sample are distributed irregular in particles. Only Fe get shown on the surface of particles.

• Journal of Plant Biology
• /
• v.3 no.2
• /
• pp.6-18
• /
• 1960

The present experiments were designed in order to clarify the differences between the long and short styled plants and the transgressive gradition in the degree of dimorphism among the three heterostylous species of the Polygonus, P. japonica, F. esculentum, and P. senticosa, based on investigations regarding the floral structure, ecological and physiological traits, the results of which are summarized as follows: (1) P. japonica, although it exhibits typical dimorphism, has undergone so high a differentiation between long and short styled that its long styled individuals behave as if they were female; and short styled individuals as if male. In long-styled individuals, filament, anther, and pollen grains show signs of degeneration, most of the pollen being abortive. On the other hand, in short styled individuals, the filament, anther, and pollen grains have attained remarkable development; the pollen grians are large and fertile. In short-plant the fertilized flowers readily drop off in every stage of their embryo development. This species has completely lost the self-fertile property, which is characteristic of the non-dimorphic Polygonum genus. Although this specsei typically exhibits the physiological characteristics of the non-dimorphic Polygonum genus. Although this specisei typically exhibits the physiological characteristics of dimorphism in controlled pollination, the short-styled individuals bear no seed in nature, thus misleading taxonomists to idenfity the short-styled plant as male. 2) The morphological feature of the flower organ of P. senticosa obviously indicates definite dimorphism. Physiologically, however, no differentiation towards dimorphism was observed, the species still retaining, both in long and short-individuals, the self-fertile property common to the Polygonum genus. Elaborate examinations revealed that regardless of the modes of pollination, both fertiization and seed setting flourish, no differentiation betwen legitimate and illegitimate unions being recognizable. This sort of physiological property has not been observed in the investigations of other heterostylous plants. It is assumed that this species is differentiated structurally into dimorphism, but not yet physiologically. In nature, however, this plant would have more opportunities to be cross-pollinated, i.e., legitimately combined, than self-pollinated because of the development of two forms of flowers. 3) In terms of heterostylism, the F. esculentum just occupies the intermediate position between P. japonica and P. senticosa structurally, ecologically, and physiologically. Doescription of some of the physiological behavior of the plant will suffice to demonstrate the above facts. While P. japonica has completely lost its self-fertile property, P. senticosa still retains it wolly. In F. esculentum 2-6% of self-fertility is the result in illegitimate combination. There occur occasionally hereditary self fertile individuals among some of the F. or 20 min. irradiation plot, when they reach any stage of the same bacterial population. In addition to this increase of total population in the plots with the more dose of UV light irradiation, it seems that the more dose of UV light irradiation is the more shortened the generation time of Azotobacter. Therefore, it is clear that variation of reproductive rate must be, mere or less, due to the genetic effects induced by UV light irradiation. On the other hand, the lag phase or logarithmic growth phase in nonirradiated culture is shortened prominently, and this must be due to the difference in bacterial number of the original inoculm. The generation time of Azotobacter is shortened by exogeneous treatment of nuclei acid derivatives, and the degree is greater in case of DNA derivatives than RNA dervatives. W.H. Price reported that the rate of ribose nucleic acid to protein in Staphylococcus muscae is proportional to the generation time: that is the faster the cell can form ribose nucleic acid, the more rapid its growth. This explains the shortening of generation time by exogeneous RNA derivatives in this work reasonably. On the other hand, it is well known that the desoxyribose nuclic acid content per cell is constant and independent of the generation time. A.D. Laren and W.N. Takahashi reported that the infectious RNA from TMV is 6 times as sensitive to inactivation by UV as it is in the form of intact virus, and that inactivation of infectious TMV involves onlu a local change on RNA chain. But, the effect of exogeneous DNA in this work suggests that irradiated living cell which cotain DNA bring about some change on DNA moleculs as well as RNA molecules. And if the mutagenic effects of UV take into consideration, it is very reasonable. Therefore, it is clear that the variation of the generation time by UV irradiation is, more or less, due to the genetic effects. Therefore, it seems that the shortness of the average lifewpan of Azotobacter by UV irradiation is resulted not only from the influence of the environmental conditions, but also from the variation of genetic factor of the individual.

• Journal of the East Asian Society of Dietary Life
• /
• v.18 no.3
• /
• pp.311-321
• /
• 2008

In order to improve the insufficient protein content of Jeung-Pyun with respect to sitology, as well as its fermentation process, this study prepared Jeung-Pyun doughs containing 0, 5, 10, 15, and 20 g of milk, respectively, and examined their mechanical and functional characteristics. The internal structure of the Jeung-Pyun was observed by SEM. With regard to color, the L-value increased with increasing amounts of added milk, and the highest value (65.43) occurred in the group containing 20 g of added milk; however, there were no significant differences among groups. The group containing 20 g of added milk also presented the highest a-value, and the samples with added milk had higher b-values than the control. At 8 hours of fermentation, the 10 g- and 15 g-added milk groups had viscosities of 5726.67 and 6600 respectively; viscosity increased with increasing amounts of milk, and the added-milk groups had significantly higher values than the control group. Hardness also increased with increasing amounts of added milk. However, there were no significant differences in resilience and cohesive power among the groups. The groups containing 5 and 10 g of added milk had the lowest cohesion and break values. For appearance, the group without added milk had the whitest color. Pore size decreased and showed less uniformity as the amount of added milk increased. The unique tackju aroma of Jeung-Pyun decreased significantly as the level of added milk increased, and the 20 g addition presented the lowest value. The 10 g-added milk group had the highest level of sweetness, and sourness increased with increasing amounts of added milk. For softness, the control group was estimated as slightly softer than the added-milk groups. The stickiest sample was generally preferred when considering taste, white color, and sweetness. Among the quality characteristics, having greater stickiness, resilience, moisture, and softness was better in terms of chewing. Overall, the size of the Jeung-Pyun containing 10 g of milk was even and well-developed.

• Korean Journal of Soil Science and Fertilizer
• /
• v.43 no.1
• /
• pp.113-118
• /
• 2010

Due to diverse soil-forming environments and different purposes of the soil classification, numerous soil classification systems have been developed worldwide. The World Reference Base for Soil Resources (WRB) and the Soil Taxonomy of the United States are well-known in Korea. However, the German Soil Systematics based on somewhat different principles from the two former systems is little-known. The objective of this paper is therefore to give a short overview of the principles of the German Soil Systematics. The German Soil Systematics consists of a six-level hierarchical structure which comprises soil divisions, soil classes, soil types, soil subtypes, soil varieties, and soil subvarieties. Soils in Germany are firstly classified into one of four soil divisions according to the soil moist regime: terrestrial soils, semi-terrestrial soils, semi-subhydric/subhydric soils, and peats. Terrestrial soils are subdivided into 13 soil classes based on the stage of soil formation and the horizon differentiation. Semi-terrestrial soils are differentiated into four classes regarding the source of soil moist: groundwater, freshwater, saltwater, and seaside. Semi-subhydric/subhydric soils are subdivided into two classes: semi-subhydric and subhydric soils. Peats are classified into two classes of natural and anthropogenic origins. Classes can be compared to orders of the U.S. Taxonomy. Classes are subdivided into 29 soil types with regard to soil forming-processes for terrestrial soils, into 17 types with regard to the soil formation for semi-terrestrial soils, into five types with regard to the content of organic matter for semi-subhydric/subhydric soils, and also into five types with regard to peat-forming processes for peats. The soil mapping units in Germany are types, which can be additionally subdivided into ca. 220 subtypes, several thousands of varieties and subvarieties using detailed nuances of morphologic features of soil profile. Soil types can be compared to great groups of the U.S. Taxonomy.

• Clean Technology
• /
• v.24 no.1
• /
• pp.41-49
• /
• 2018

Y zeolites with different extra-framework cations, such as $Na^+$, $N^+$, $Ca^{2+}$, and $Cu^{2+}$, with different charge and ionic radius have been investigated to greatly enhance a working capacity (W) of $CO_2$ adsorption at $25^{\circ}C$ and a $CO_2/CO$ selectivity factor (S). A sample of NaY with a very small amount of 0.012% $Ca^{2+}$ was fully reversible for seven times repeated $CO_2$ adsorption/desorption cycles, thereby forming no surface carbonates unlikely earlier reports. Although at pressures above 4 bar, 2.00% CaY, 1.60% CuY and 1.87% LiY all showed a $CO_2$ adsorption very similar to that measured for NaY, they gave a significant decrease in the adsorption at lower pressures, depending on the metal ion. At 0.5 ~ 2.5 bar, the extent of $CO_2$ adsorption was in the order NaY > 1.60% CuY > 2.00% CaY > 1.87% LiY. All the $Na^+-based$ metals-exchanged zeolites have a FAU (faujasite) framework and a Si/Al value near 2.6; thus, there is no discernible difference in the framework topology, framework chemical compositions, effective aperture size, and channel structure between the zeolite samples. Therefore, the distinctive behavior in the adsorption of $CO_2$ with a character as a weak Lewis acid is associated with the site basicity of the zeolites, and the interaction potentials of the cations. Different trend was shown for a CO adsorption due to weaker quadrupole interactions. Adsorption of $CO_2$ and CO on samples of CaY with 0.012 to 5.23% Ca disclosed a significant dependence on the Ca loading. The $CO_2$ adsorption increased when the cation exists up to ca. 0.05%, while it decreased at higher Ca amounts. However, values for both W and S could greatly increase as the bare zeolite is enriched by $Ca^{2+}$ ions. The 5.23% CaY had $W=2.37mmol\;g^{-1}$ and S = 4.37, and the former value was comparable to a benchmark reported in the literature.

• Korean Journal of Agricultural Science
• /
• v.10 no.1
• /
• pp.9-21
• /
• 1983

The structures of oak stands were studied in relation to their productivity. Three plots classified by the site quality following to altitude, were selected on the oak stand located at Naryongri Eunsanmyun Buyeogun. The site qualities of the stand studied were generally estimated to be moderate in respect to some soil physical and chemical properties. Quercus variabilis was dominant species in the vegetation composition of the stand studied. The crown densities of the stand were varied from 65.4% in plot I to 78.2% in plot III and the parts, occupied with oak trees, ranged from 44.4% in plot I to 65.9% in plot III in the density. In the contrast to crown density, the growing stocks of oak trees ranged from $3.937m^3$ in plot I (73.4% of plot total) to $2.075m^3$ in plot III (84.3% of plot total). The occupied ratios, measured by crown class, exhibited dissimilarity between crown projection area and volume, and also the ratios brought into different patterns by plot. Highly significant correlations were proved between crown projection areas and volumes of individual oak trees by plot, but not proved in the relations by crown class. The cumulative growth, current annual increment and mean annual increment displayed various patterns by plot and by crown class in the heights, D. B. H. s and volumes of oak trees. The maximum values of current annual increment of height were generally recorded in 10 to 20 years, earlier than those of D. B. H. The mean annual increment of height, D. B. H. and volume maintained similar levels after about 10 years and fluctuated irregularly. The crown projection area and volume following to thinning decreased in the order of the thinning methods of grade A < grade B < crown thinning$90m^3/ha$ in about 40 years in the productivity of volume of the oak stand studied.

• Journal of Dental Rehabilitation and Applied Science
• /
• v.28 no.4
• /
• pp.359-370
• /
• 2012

The endodontically treated tooth is generally restored with post and core, owing to the brittle and the loss of large amount of tooth structure. The purpose of this study was to evaluate the clinical status of fixed prostheses to improve the quality of dental care. In order to assess the clinical status of fixed prostheses, a total of 101 individuals (aged 30-89, 66 women and 35 men loaded with 125 fixed prostheses) who treated in the Department of Prosthodontics, Pusan National University Dental Hospital, between January 1990 to December 2005 were examined. The results of this study were as follows: 1. Length of service of fixed prostheses was $9.7{\pm}3.4$ years (mean), 11.1 years (median). 2. Age and sex of patient was found to have no statistically significant influence on longevity of fixed prostheses (P>.05). 3. Location of fixed prostheses was found to have statistically significant influence on longevity of fixed prostheses (P<.05). The longevity of fixed prostheses was low in anterio-posterior combination region (median:9.2 years). 4. Longevity of fixed prostheses made of base metal ceramic(median:12.0 years) and noble metal ceramic (median:11.3 years) is long (P<.05). 5. Number of units in fixed prostheses was found to have no statistically significant influence on longevity of fixed prostheses (P>.05). 6. Condition of opposing dentition was found to have no statistically significant influence on longevity of fixed prostheses (P>.05). 7. Dental caries, periapical disease, tooth fracture were frequent complications. In 51.9% of the cases, abutment state after removing fixed prostheses was needed to be extracted.

• Sijohaknonchong
• /
• v.43
• /
• pp.41-66
• /
• 2015

A study on rhythm of a sijo was mostly conducted based on rhythm theory. As it is considered to define the rhythm of a formal sijo based on three verses, its significance has been recognized. However, if rhythm is understood to be superior to cadence or versification, it seems necessary to examine the rhythm of a sijo as a verse with a fixed form as well as a highly individual rhythm of each and every lyric poet, which is informal rhythm, in order to fully understand them. In this case, prosodie analysis by H. Meschonnic (1932~ 2009) can be a significant methodology. As this study gropes for a possibility to examine the rhythm of a sijo from a new perspective instead of existing rhythm theory through the application of H. Meschonnic's prosodie analysis, it can be regarded as an essay. Prosodie newly suggested by Meschonnic is referred to as linguistic organization of consonants and vowels and indication of their paradigm, and it conflicts the perspective that traditionally separates linguistic sound from meaning for dichotomous understanding. It is due to the fact that the organization of consonants and vowels is a unit that constitutes a complicated layer of significant sound and meaning. Accordingly, prosodie analysis that is irregularly and aperiodically distributed within poetic text can be considered as methodology aimed at explaining how a poem is integrated in terms of sound and semantics. The core of prosodie analysis is to examine how the phonologic system stands against the theme of a poem. It ultimately has the same way of establishing literary style of a poet as it is to explain a unique aesthetic structure that individual poems have and show distinct characteristics of linguistic use by a poet. Prior to application of the prosodie analysis to sijo in general, the study preparatorily conducted prosodie analysis on < Ouga > by Gosan Seon-do Yun.

• Journal of Service Research and Studies
• /
• v.10 no.3
• /
• pp.23-42
• /
• 2020

This study conceptualized the essence and propensity of modern management in service age, based on philosophy, and developed items to evaluate the conceptualized content. It was carried out as a new study to deepen the study of management philosophy and management theory by the new management framework. In order to establish the philosophical foundation of the modern management, the essence of the modern management was conceptualized based on the fundamental ideas of the East and West, and then an evaluation item was developed to put the essence and propensity of the modern management into practical use through analytical and empirical methods. After analyzing the representative ideas of mankind, it was derived that the Book of Change has the qualification as a philosophical model that can derive the essence of modern management. The Book of Change explains the reasoning of the world in the structure of two opposing parties, such as Taiji or Yin and Yang, and the process of acknowledging the contradictions within each opposing party and overcoming the contradictions through change is the central idea. Because you can see. After conducting a conceptual study, through empirical research, the essence and propensity of a modern manager should be conceptualized. The concept of essence and empirical study of the modern management using the leading role was conducted in two stages. First, a qualitative study using repetitive comparative analysis (CCM), focus group interview (FGI), and text mining was conducted to derive the essential and propensity conceptualization items that modern managers should possess. In addition, a quantitative study using factor analysis to develop sample items and develop measurement items through literature review and FGI was conducted to derive the essential concept of the modern management. Finally, the essence of modern management was derived: learning, preparation, challenge, inclusion, trust, morality, and sacrifice. In the future, it is necessary to conduct empirical research on the effectiveness of the essence of modern management for global and Korean representative companies.