• Journal of Intelligence and Information Systems
• /
• v.25 no.1
• /
• pp.219-239
• /
• 2019

As the importance of providing customized services to individuals becomes important, researches on personalized recommendation systems are constantly being carried out. Collaborative filtering is one of the most popular systems in academia and industry. However, there exists limitation in a sense that recommendations were mostly based on quantitative information such as users' ratings, which made the accuracy be lowered. To solve these problems, many studies have been actively attempted to improve the performance of the recommendation system by using other information besides the quantitative information. Good examples are the usages of the sentiment analysis on customer review text data. Nevertheless, the existing research has not directly combined the results of the sentiment analysis and quantitative rating scores in the recommendation system. Therefore, this study aims to reflect the sentiments shown in the reviews into the rating scores. In other words, we propose a new algorithm that can directly convert the user 's own review into the empirically quantitative information and reflect it directly to the recommendation system. To do this, we needed to quantify users' reviews, which were originally qualitative information. In this study, sentiment score was calculated through sentiment analysis technique of text mining. The data was targeted for movie review. Based on the data, a domain specific sentiment dictionary is constructed for the movie reviews. Regression analysis was used as a method to construct sentiment dictionary. Each positive / negative dictionary was constructed using Lasso regression, Ridge regression, and ElasticNet methods. Based on this constructed sentiment dictionary, the accuracy was verified through confusion matrix. The accuracy of the Lasso based dictionary was 70%, the accuracy of the Ridge based dictionary was 79%, and that of the ElasticNet (${\alpha}=0.3$) was 83%. Therefore, in this study, the sentiment score of the review is calculated based on the dictionary of the ElasticNet method. It was combined with a rating to create a new rating. In this paper, we show that the collaborative filtering that reflects sentiment scores of user review is superior to the traditional method that only considers the existing rating. In order to show that the proposed algorithm is based on memory-based user collaboration filtering, item-based collaborative filtering and model based matrix factorization SVD, and SVD ++. Based on the above algorithm, the mean absolute error (MAE) and the root mean square error (RMSE) are calculated to evaluate the recommendation system with a score that combines sentiment scores with a system that only considers scores. When the evaluation index was MAE, it was improved by 0.059 for UBCF, 0.0862 for IBCF, 0.1012 for SVD and 0.188 for SVD ++. When the evaluation index is RMSE, UBCF is 0.0431, IBCF is 0.0882, SVD is 0.1103, and SVD ++ is 0.1756. As a result, it can be seen that the prediction performance of the evaluation point reflecting the sentiment score proposed in this paper is superior to that of the conventional evaluation method. In other words, in this paper, it is confirmed that the collaborative filtering that reflects the sentiment score of the user review shows superior accuracy as compared with the conventional type of collaborative filtering that only considers the quantitative score. We then attempted paired t-test validation to ensure that the proposed model was a better approach and concluded that the proposed model is better. In this study, to overcome limitations of previous researches that judge user's sentiment only by quantitative rating score, the review was numerically calculated and a user's opinion was more refined and considered into the recommendation system to improve the accuracy. The findings of this study have managerial implications to recommendation system developers who need to consider both quantitative information and qualitative information it is expect. The way of constructing the combined system in this paper might be directly used by the developers.

• Applied Biological Chemistry
• /
• v.17 no.1
• /
• pp.1-30
• /
• 1974

Present study was undertaken to elucidate the relationship between uptake of nutrients and photosynthetic activities, and the translocation of several mineral nutrients in rice plants which were grown under different cultural conditions, utilizing radioactive tracer technique. Particular emphasis was placed on the analysis of patterns of nutrient uptake, the relationship between nutritional conditions and yield components. For this, rice plants grown on either low or high yielding fields at different growth stage were subjected to this study. The results are summarized as follows; 1. Varietal difference was observed in the uptake of potassium and phosphorus. Kusabue and Jinheung had good capacity but Paldal had rather poor capacity for the uptake of the both nutrients. 2. For rice plants, a high positive correlation was found between the oxidation of alpha plaus-naphthylamine by root and uptake of phosphorus. 3. Carbon assimilation rate repended on rice varieties. It was high in Noindo, Gutaenajuok #3 Suweon #82 and Jinheung but low in Taegujo, Kwanok, Yugu #132 etc. 4. Heavy application of nitrogen increased carbon assimilation in rice plants but this also depressed translocation of certain carbohydrates to ears. 5. Carbon assimilation wan greatly hampered in rice plants deficient in magnesium, phosphorus or potassium. 6. Total dry matter after ear formation stage, was much higher in rice plants grown in high yielding fields than those grown in low yielding fields. 7. Leaf area index(LAI) reached maximum at heading stage and decreased thereafter in high yielding fields. But in low yielding fields, it reached maximum before heading and sharply decreased thereafter due to early senescence of lower leaves. 8. In general, light transmission ratio (LTR) of leaves was higher in the early growth stage and lower in later stages. Higher ratio of LTR to leaf area index, was found in the rice grown in high yielding fields than those in low yielding fields. 9. Net photosynthetic activity decreased with the increase in leaf area index but was higher in high yielding fields than in low yielding fields. 10. After the ear formation stage, nitrogen, potassium and silicon as weil as $K_2O/N$ in straw were higher in high yielding fields than those in low yielding fields. 11. Nitrogen, phosphorus, potassium and magnesium taken up by rice plants in low yielding fields before heading stage were readily translocated to ears than those in high yielding fields. This suggests greater redistribution of nutrients in straw occurs due to lower uptake, in later growth stages, by rice plants grown in low yielding fields and hence results in early senescence due to nutrient deprivation. 12. In the high yielding fields nitrogen uptake by rice was slow but continuous throughout the life of the plants resulting in a large uptake even after heading. But, in low yielding fields the uptake was fast before heading and slow after heading. 13. A high positive correlation was found between the contents of nitrogen and potassium in the straw at heading stage and grain yield. Positive correlation was also found to hold between the contents of potassium, silicon, $K_2O/N$, $SiO_2/N$ in the straw at harvesting stage, and grain yield. 14. Carbon assimilation was greately hampered in rice plants deficient in magensium, phosphorus or potassium. 15. Uptake of nitrogen, phosphorus, potassium, silicon and manganese by rice was considerably higher in high yielding fields and reached maximum at ear formation stage. 16. In rice, a high positive correlation was discovered between total uptake of nitrogen, phosphorus, potassium, calcium, magnesium, silicon, manganese at harvesting stage and grain yield. 17. In rice, a high positive correlation was found between the total uptake of nitrogen, phosphorus, potassium, calcium, magnesium, silicon at harvesting stage, and number of spikelets per $3.3\;m^2$. In addition, a correlation was found between the total uptake of nitrogen and potassium and number of panicles per hill.

• Journal of Intelligence and Information Systems
• /
• v.18 no.4
• /
• pp.59-77
• /
• 2012

This study focuses on predicting which firms will increase capital by issuing new stocks in the near future. Many stakeholders, including banks, credit rating agencies and investors, performs a variety of analyses for firms' growth, profitability, stability, activity, productivity, etc., and regularly report the firms' financial analysis indices. In the paper, we develop predictive models for rights issues using these financial analysis indices and data mining techniques. This study approaches to building the predictive models from the perspective of two different analyses. The first is the analysis period. We divide the analysis period into before and after the IMF financial crisis, and examine whether there is the difference between the two periods. The second is the prediction time. In order to predict when firms increase capital by issuing new stocks, the prediction time is categorized as one year, two years and three years later. Therefore Total six prediction models are developed and analyzed. In this paper, we employ the decision tree technique to build the prediction models for rights issues. The decision tree is the most widely used prediction method which builds decision trees to label or categorize cases into a set of known classes. In contrast to neural networks, logistic regression and SVM, decision tree techniques are well suited for high-dimensional applications and have strong explanation capabilities. There are well-known decision tree induction algorithms such as CHAID, CART, QUEST, C5.0, etc. Among them, we use C5.0 algorithm which is the most recently developed algorithm and yields performance better than other algorithms. We obtained data for the rights issue and financial analysis from TS2000 of Korea Listed Companies Association. A record of financial analysis data is consisted of 89 variables which include 9 growth indices, 30 profitability indices, 23 stability indices, 6 activity indices and 8 productivity indices. For the model building and test, we used 10,925 financial analysis data of total 658 listed firms. PASW Modeler 13 was used to build C5.0 decision trees for the six prediction models. Total 84 variables among financial analysis data are selected as the input variables of each model, and the rights issue status (issued or not issued) is defined as the output variable. To develop prediction models using C5.0 node (Node Options: Output type = Rule set, Use boosting = false, Cross-validate = false, Mode = Simple, Favor = Generality), we used 60% of data for model building and 40% of data for model test. The results of experimental analysis show that the prediction accuracies of data after the IMF financial crisis (59.04% to 60.43%) are about 10 percent higher than ones before IMF financial crisis (68.78% to 71.41%). These results indicate that since the IMF financial crisis, the reliability of financial analysis indices has increased and the firm intention of rights issue has been more obvious. The experiment results also show that the stability-related indices have a major impact on conducting rights issue in the case of short-term prediction. On the other hand, the long-term prediction of conducting rights issue is affected by financial analysis indices on profitability, stability, activity and productivity. All the prediction models include the industry code as one of significant variables. This means that companies in different types of industries show their different types of patterns for rights issue. We conclude that it is desirable for stakeholders to take into account stability-related indices and more various financial analysis indices for short-term prediction and long-term prediction, respectively. The current study has several limitations. First, we need to compare the differences in accuracy by using different data mining techniques such as neural networks, logistic regression and SVM. Second, we are required to develop and to evaluate new prediction models including variables which research in the theory of capital structure has mentioned about the relevance to rights issue.

• Journal of the Korean Society of Groundwater Environment
• /
• v.7 no.3
• /
• pp.103-115
• /
• 2000

The formation of brown-colored precipitates is one of the serious problems frequently encountered in the development and supply of groundwater in Korea, because by it the water exceeds the drinking water standard in terms of color. taste. turbidity and dissolved iron concentration and of often results in scaling problem within the water supplying system. In groundwaters from the Pajoo area, brown precipitates are typically formed in a few hours after pumping-out. In this paper we examine the process of the brown precipitates' formation using the equilibrium thermodynamic and kinetic approaches, in order to understand the origin and geochemical pathway of the generation of turbidity in groundwater. The results of this study are used to suggest not only the proper pumping technique to minimize the formation of precipitates but also the optimal design of water treatment methods to improve the water quality. The bed-rock groundwater in the Pajoo area belongs to the Ca-$HCO_3$type that was evolved through water/rock (gneiss) interaction. Based on SEM-EDS and XRD analyses, the precipitates are identified as an amorphous, Fe-bearing oxides or hydroxides. By the use of multi-step filtration with pore sizes of 6, 4, 1, 0.45 and 0.2 $\mu\textrm{m}$, the precipitates mostly fall in the colloidal size (1 to 0.45 $\mu\textrm{m}$) but are concentrated (about 81%) in the range of 1 to 6 $\mu\textrm{m}$in teams of mass (weight) distribution. Large amounts of dissolved iron were possibly originated from dissolution of clinochlore in cataclasite which contains high amounts of Fe (up to 3 wt.%). The calculation of saturation index (using a computer code PHREEQC), as well as the examination of pH-Eh stability relations, also indicate that the final precipitates are Fe-oxy-hydroxide that is formed by the change of water chemistry (mainly, oxidation) due to the exposure to oxygen during the pumping-out of Fe(II)-bearing, reduced groundwater. After pumping-out, the groundwater shows the progressive decreases of pH, DO and alkalinity with elapsed time. However, turbidity increases and then decreases with time. The decrease of dissolved Fe concentration as a function of elapsed time after pumping-out is expressed as a regression equation Fe(II)=10.l exp(-0.0009t). The oxidation reaction due to the influx of free oxygen during the pumping and storage of groundwater results in the formation of brown precipitates, which is dependent on time, $Po_2$and pH. In order to obtain drinkable water quality, therefore, the precipitates should be removed by filtering after the stepwise storage and aeration in tanks with sufficient volume for sufficient time. Particle size distribution data also suggest that step-wise filtration would be cost-effective. To minimize the scaling within wells, the continued (if possible) pumping within the optimum pumping rate is recommended because this technique will be most effective for minimizing the mixing between deep Fe(II)-rich water and shallow $O_2$-rich water. The simultaneous pumping of shallow $O_2$-rich water in different wells is also recommended.

• Investigative Magnetic Resonance Imaging
• /
• v.16 no.1
• /
• pp.16-24
• /
• 2012

Purpose : The purpose of this study was to determine whether there is a significant correlation between vertebral marrow fat fraction measured using Dixon quantitative chemical shift MRI (QCSI) and BMD on dual-energy X-ray absorptiometry (DXA). Materials and Methods: This retrospective study included 68 healthy individuals [mean age, 50.7 years; range, 25-76; male/female (M/F) = 36/32] who underwent DXA of the L-spine and whole body MRI including QCSI of the L-spine and chemical shift MRI of the liver. The enrolled individuals were divided into subgroups according to sex and T-score [i.e., normal bone density (M/F=27/23) and osteopenia (M/F=9/9)]. Vertebral marrow (Dixon QCSI, TR/TE 10.2/4.8 ms) and hepatic fat fractions (chemical shift technique, TR/TE 110/4.9 and 2.2 ms) were calculated on MRI. We evaluated whether there were significant differences in age, body mass index (BMI), vertebral marrow fat fraction, or hepatic fat fraction among the subgroups. Whether or not the participant had reached menopause was also evaluated in females. The correlations among variables (i.e., age, BMI, vertebral marrow and hepatic fat fractions, BMD) were evaluated using Spearman's correlation method. Results: There were no significant differences in age, BMI, or vertebral marrow and hepatic fat fractions between the two male subgroups (normal bone density vs. osteopenia). In female subjects, mean age in the osteopenic subgroup was greater than that in the normal subgroup (p=0.01). Presence of menopause was more common in the osteopenic subgroup [77.8% (7/9)] than the normal subgroup [26.1% (6/23), p<0.05]. The other variables showed no significant difference between female subgroups. The only significant correlation with marrow fat fraction after partial correlation analysis was that with age in the female subjects (r=0.43, p<0.05). Conclusion: The vertebral marrow fat fraction calculated using the Dixon QCSI does not precisely reflect the mild decrease in BMD for either sex.

• Restorative Dentistry and Endodontics
• /
• v.27 no.6
• /
• pp.577-586
• /
• 2002

The purpose of this in vitro study was to compare the effects of root canal cleanness following two Ni-Ti rotary instruments with different rake angle. Thirty-six sound, extracted human premolars with single root were randomly divided into three groups. The used rotary instruments were HEROShaper (Group 1, Micro-Mega, Besancon, France, n=12) and ProFile (Group 2, Maillefer, Ballaigues, Switzerland, n=12). Control group (n=12) was only extirpated with barbed broach (Mani, Matsutani Seisakusho Co., Japan) Group 1 & 2 teeth were prepared to a #40/.04 taper at the apex followed by 1 mm using crown-down technique. After canal preparation and frequent irrigation with 5.25% sodium hypochlorite, the roots split longitudinally into a bucco-lingual direction. Root halves were cross-sectioned in apical third portion again. All root specimens were processed for SEM investigation and photographed. Separate evaluations by one endodontist were undertaken for smear layer on prepared walls with a five score-index for each using reference photograph in root halves. The penetration depth of smear layer into dentinal tubules was also estimated in the other halves. Following results were obtained: 1. Smear layer was observed on all the prepared walls with two experimental groups except control group. 2. Smear layer characteristics in two experimental groups; 1) HEROShaper group showed snowy, dusty appearance and were shown open dentinal tubuli on the prepared walls of almost specimens, and the thickness of smear layer covering onto dentinal surfaces was within 1-2 ${\mu}m$ in a few specimens. 2) ProFile group showed shiny, burnished appearance and complete root canal wall covered by a homogenous smear layer with no open dentinal tubuli in all specimens. The penetration of smear layer into dentinal tubules was found in all specimens and the thickness was at 2-4 ${\mu}m$ in all specimens. These results demonstrated that a completely clean root canal could not be achieved regardless of positive or negative rake angle, which is in accordance with the majority of previous studies on root canal cleanliness In conclusion, through irrigation with antibacterial solutions or chelating agents is recommended to remove the smear layer on prepared canal wall in spite of Ni-Ti instrumentation.

• Korean Journal of Weed Science
• /
• v.3 no.1
• /
• pp.69-74
• /
• 1983

The behaviour of mixtures of herbicides was determined to obtain the basic informations about effective herbicide use, enhancing herbicidal efficacy and reducing the chemical cost. Fourteen herbicides with 91 mixed combinations were evaluated by Limpel et al method at the Echinochloa crus galli Beauv-Monochuria vaginalis Presl.-Scirpus hotarui Ohwi (importance values of these weeds were 63%, 16% and 10%, respectively) community type. Thirty eight mixed combinations showed the antagonistic response. Among these 14 mixed combinations including chlormethoxynil + naproanilide mixture were greater than 11% in antagonistic effect. On the other hand, 40 mixed combinations including chlormethoxynil + SW751 mixture showed additive response (${\pm}2%$). For synergistic response, 13 mixed combinations were belonged to this group. Particularly, 3 mixed combinations, chlormethoxynil + butachlor, chlormethoxynil + bifenox and nitrofen + ACN/MCPB/nitrofen mixtures were greater than 11% in synergistic effects. The mixture of thiobencarb + oxyfluorfen was analyzed by isobole technique. This mixture showed the synergistic response and the interaction index was approximately 2. The most optimum mixtur for inducing 90%n weed suppression was 0.012 kg ai/ha for oxyfluorfen and 0.45 kg ai/ha for thiobencarb.

• Journal of Dairy Science and Biotechnology
• /
• v.33 no.4
• /
• pp.263-269
• /
• 2015

Various carbohydrates (lactose, glucose, and fructose), lactic acid, uric acid, and acetoin were separated on a ZORBAX Carbohydrate Analysis column using the Agilent 1200 HPLC ChemStation$^{TM}$, and were identified according to retention times with 325 Dual Wavelength UV-Vis Detector and Refractive Index Detector with 0.013 N $H_2SO_4$ at a flow rate of 0.8 mL/min. In addition, the lactase activity of four commercial probiotic lactic acid bacteria during 6-hour incubation was determined using a high-performance liquid chromatography (HPLC) method. Among the tested samples, Bifidobacterium animalis subsp. lactis showed the greatest lactase activity, followed by Lactobacillus rhamnosus and Lactobacillus casei, with Streptococcus salivarius subsp. thermophilus showing the lowest activity. Therefore, this HPLC technique shows potential for evaluating the fermentation processes of probiotic lactic acid bacteria and could simultaneously confirm the degree of ripening in various fermented dairy foods within only a half hour.

• Current Research on Agriculture and Life Sciences
• /
• v.8
• /
• pp.129-138
• /
• 1990

This study investigated the relationship between the cutflow of rural population and agricultural policy by using time series method. For the analytical tools, decomposition time series methods and regression technique were employed in computing seasonal fluctuation and cyclical fluctuation of population migration. Also, this study predicted farmhouse, rural population till the 2000's by means of the mathematical methods. The analytical forms employed in forecasting farmhouse, rural population were Exponential curve, Gompertz curve and Transcendental form. The major findings of this study were identified as follows: 1) Rural population and farmhouse population began to decrease from 1965 and hastily went down since 1975. Rural population which accounted for 36.4 percent, 35.6 percent of national population respectively in 1960 diminished about two times: 17.5 percent, 17.1 percent respectively. 2) The rapid decreasing of the rural population was caused because of the outflow of rural people to the urban regions. Of course, that was also caused from the natural decreases but the main reason was heavily affected more the former than the latter. In the outflowing course shaped from rural to the urban regions, rural people concentrated on such metropolis as Seoul, Pusan, Keanggi. But these trends were diminishing slowly. On the other hand, compared with that of the 1970's the migration to Keanggi was still increasing in the 1980's. That is, people altered the way of migration from the migration to Seoul, Pusan to the migration to the out-skirts of Seoul. 3) The seasonal fluctuation index of population migration has gone down since the June which the request of agricultural labor force increases and has turned to be greatly wanted in the March as result of decomposition time series method. As result of cyclical analysis, the cyclical patterns of migration have greatly 7 cycle. 4) As result of forecasting the rural and farmhouse population, rural and farmhouse population in the 2000 will be about 9,655(thousand/people) and 4,429(thousand/people) respectively. Thus, it is important to analyze the probloms that rural and farmhouse population will decrease or increase by the degree. But fairly defining the agricultural into a industry that supply the food, this problem - how much our nation need the rural and farmhouse population - is greatly significant too. Therefore, the basic problems of the agricultural including the outflows of rural people are the earning differentials between rural and urban regions. And we should regard the problems of the gap of relative incomes between rural and urban regions as the main task of the agricultural policy and treat the agricultural policy in the viewpoint of developing economic equilibrium than efficiency by using actively the natural resources of the rural regions.

• Journal of Intelligence and Information Systems
• /
• v.17 no.4
• /
• pp.241-254
• /
• 2011

Financial time-series forecasting is one of the most important issues because it is essential for the risk management of financial institutions. Therefore, researchers have tried to forecast financial time-series using various data mining techniques such as regression, artificial neural networks, decision trees, k-nearest neighbor etc. Recently, support vector machines (SVMs) are popularly applied to this research area because they have advantages that they don't require huge training data and have low possibility of overfitting. However, a user must determine several design factors by heuristics in order to use SVM. For example, the selection of appropriate kernel function and its parameters and proper feature subset selection are major design factors of SVM. Other than these factors, the proper selection of instance subset may also improve the forecasting performance of SVM by eliminating irrelevant and distorting training instances. Nonetheless, there have been few studies that have applied instance selection to SVM, especially in the domain of stock market prediction. Instance selection tries to choose proper instance subsets from original training data. It may be considered as a method of knowledge refinement and it maintains the instance-base. This study proposes the novel instance selection algorithm for SVMs. The proposed technique in this study uses genetic algorithm (GA) to optimize instance selection process with parameter optimization simultaneously. We call the model as ISVM (SVM with Instance selection) in this study. Experiments on stock market data are implemented using ISVM. In this study, the GA searches for optimal or near-optimal values of kernel parameters and relevant instances for SVMs. This study needs two sets of parameters in chromosomes in GA setting : The codes for kernel parameters and for instance selection. For the controlling parameters of the GA search, the population size is set at 50 organisms and the value of the crossover rate is set at 0.7 while the mutation rate is 0.1. As the stopping condition, 50 generations are permitted. The application data used in this study consists of technical indicators and the direction of change in the daily Korea stock price index (KOSPI). The total number of samples is 2218 trading days. We separate the whole data into three subsets as training, test, hold-out data set. The number of data in each subset is 1056, 581, 581 respectively. This study compares ISVM to several comparative models including logistic regression (logit), backpropagation neural networks (ANN), nearest neighbor (1-NN), conventional SVM (SVM) and SVM with the optimized parameters (PSVM). In especial, PSVM uses optimized kernel parameters by the genetic algorithm. The experimental results show that ISVM outperforms 1-NN by 15.32%, ANN by 6.89%, Logit and SVM by 5.34%, and PSVM by 4.82% for the holdout data. For ISVM, only 556 data from 1056 original training data are used to produce the result. In addition, the two-sample test for proportions is used to examine whether ISVM significantly outperforms other comparative models. The results indicate that ISVM outperforms ANN and 1-NN at the 1% statistical significance level. In addition, ISVM performs better than Logit, SVM and PSVM at the 5% statistical significance level.