• Journal of Intelligence and Information Systems
• /
• v.27 no.3
• /
• pp.57-73
• /
• 2021

Maintenance and prevention of failure through anomaly detection of ICT infrastructure is becoming important. System monitoring data is multidimensional time series data. When we deal with multidimensional time series data, we have difficulty in considering both characteristics of multidimensional data and characteristics of time series data. When dealing with multidimensional data, correlation between variables should be considered. Existing methods such as probability and linear base, distance base, etc. are degraded due to limitations called the curse of dimensions. In addition, time series data is preprocessed by applying sliding window technique and time series decomposition for self-correlation analysis. These techniques are the cause of increasing the dimension of data, so it is necessary to supplement them. The anomaly detection field is an old research field, and statistical methods and regression analysis were used in the early days. Currently, there are active studies to apply machine learning and artificial neural network technology to this field. Statistically based methods are difficult to apply when data is non-homogeneous, and do not detect local outliers well. The regression analysis method compares the predictive value and the actual value after learning the regression formula based on the parametric statistics and it detects abnormality. Anomaly detection using regression analysis has the disadvantage that the performance is lowered when the model is not solid and the noise or outliers of the data are included. There is a restriction that learning data with noise or outliers should be used. The autoencoder using artificial neural networks is learned to output as similar as possible to input data. It has many advantages compared to existing probability and linear model, cluster analysis, and map learning. It can be applied to data that does not satisfy probability distribution or linear assumption. In addition, it is possible to learn non-mapping without label data for teaching. However, there is a limitation of local outlier identification of multidimensional data in anomaly detection, and there is a problem that the dimension of data is greatly increased due to the characteristics of time series data. In this study, we propose a CMAE (Conditional Multimodal Autoencoder) that enhances the performance of anomaly detection by considering local outliers and time series characteristics. First, we applied Multimodal Autoencoder (MAE) to improve the limitations of local outlier identification of multidimensional data. Multimodals are commonly used to learn different types of inputs, such as voice and image. The different modal shares the bottleneck effect of Autoencoder and it learns correlation. In addition, CAE (Conditional Autoencoder) was used to learn the characteristics of time series data effectively without increasing the dimension of data. In general, conditional input mainly uses category variables, but in this study, time was used as a condition to learn periodicity. The CMAE model proposed in this paper was verified by comparing with the Unimodal Autoencoder (UAE) and Multi-modal Autoencoder (MAE). The restoration performance of Autoencoder for 41 variables was confirmed in the proposed model and the comparison model. The restoration performance is different by variables, and the restoration is normally well operated because the loss value is small for Memory, Disk, and Network modals in all three Autoencoder models. The process modal did not show a significant difference in all three models, and the CPU modal showed excellent performance in CMAE. ROC curve was prepared for the evaluation of anomaly detection performance in the proposed model and the comparison model, and AUC, accuracy, precision, recall, and F1-score were compared. In all indicators, the performance was shown in the order of CMAE, MAE, and AE. Especially, the reproduction rate was 0.9828 for CMAE, which can be confirmed to detect almost most of the abnormalities. The accuracy of the model was also improved and 87.12%, and the F1-score was 0.8883, which is considered to be suitable for anomaly detection. In practical aspect, the proposed model has an additional advantage in addition to performance improvement. The use of techniques such as time series decomposition and sliding windows has the disadvantage of managing unnecessary procedures; and their dimensional increase can cause a decrease in the computational speed in inference.The proposed model has characteristics that are easy to apply to practical tasks such as inference speed and model management.

• Journal of Dairy Science and Biotechnology
• /
• v.26 no.1
• /
• pp.1-4
• /
• 2008

Melatonin is one of the functional hormones in the milk and found in all mammalian species. The primary motivation for the use of melatonin as a supplement is as a natural aid to better sleep. Melatonin contents in milk vary according to the naturally occurring factors such as seasons and milking times. As a number of studies indicate melatonin supplementation helps to reduce the age-related decline in hormone production and overall enhancement of health, the more detailed studies are required to know the nature of milk derived hormones and to clarify them as a functional value adaptation. The objectives of this study were to examine the change of melatonin content in bovine milk during different milking times at the two dairy farms and during dairy process including homogenization and pasteurization. Commercial dairy products including night time milking products(night milk) marketed in Japan were also investigated. Melatonin content was determined by radioimmunoassay using 125I. Individual milk was collected from lactating Holstein cows at Kyong-Ki province in Korea. At farm A, the melatonin contents milking at 4 a.m. was higher than those of 7 p.m. and 12 p.m.($6.90{\pm}3.55,\;2.01{\pm}1.47$ and $0.16{\pm}0.04pg/mL$, respectively). At farm B, the mean melatonin contents milked from 24:00 AM to 05:00 AM have shown the highest contents ($4.65{\pm}$0.72pg/mL), and milking samples of the rest time had very little or non-detection of melatonin contents. Melatonin contents of commercial night milk samples were about 5 times higher than those of other common milk products. These results suggested that melatonin might be heat-resistant and commercial dairy process might not affect its contents in the final products.

• Applied Chemistry for Engineering
• /
• v.21 no.4
• /
• pp.372-376
• /
• 2010

Non-conventional energy is considered as important future energy source, as conventional energy has limitation for its capacity. The demand on value added process in heavy oil/oil sand bitumen is increasing in particular. Solvent Deasphalting (SDA) process for Deasphalted Oil (DAO) is used as heavy oil upgrading process in existing refinery process. SDA process for heavy oil upgrading has been already commercialized by leading countries. SDA R&D projects have been carried out actively by those countries. In this study, patent analysis for SDA technology development was carried out. From 1970's, when SDA patents were applied, the patents in Korea, USA, Japan, Canada and Europe were searched and distributed to extraction, recovery, solvent and etc. 334 patents were selected relating to heavy oil upgrading SDA process. The application status of SDA process patents showed a tendency to increase slightly. The number of patent applied was USA patent 131 (39%), Canada patent 83 (25%), Japan patent 35 (11%) and Korea patent 6 (2%). It will be necessary for efficient use of energy resource to support SDA R&D by government.

• Asian-Australasian Journal of Animal Sciences
• /
• v.8 no.3
• /
• pp.281-287
• /
• 1995

A study was conducted to investigate the effect of different inclusion levels of urea treated whole-crop wheat silage (UWCWS) in grass silage based rations on the performance of growing beef cattle. The winter wheat (variety, Riband) was harvested (in the summer of 1991) at a dry matter proportion of 520 g/kg and treated with feed grade urea at the rate of 37 kg/tonne crop dry matter and preserved in a heavy duty plastic bag using a silo press. The urea treated whole crop wheat silage (UWCWS) was mixed with grass silage to replace 0.00 (S100), 0.33 (S33) and 0.67 (S67) parts of the forage dry matter and fed ad libitum in a cross over design to 18 Simmental X Holstein Friesian growing beef animals. Two energy sources {one high in starch, rolled barley (RB) and one high in digestible fibre, sugar beet pulp (SBP)} were fed to supply sufficient energy for the efficient use of nitrogen by the rumen micro-organisms. The data on DMIF (dry matter intake of forage), TDMI (total dry matter intake), DLWG (daily live weight gain), FCR (feed conversion ratio) were recorded and faecal samples were collected to determine the digestibility coefficients. Results revealed that with the inclusion of UWCW in the animals' diets the DMI of the forage was significantly increased (p < 0.05). The highest DMIF was found in the treatment "S33" ($6.28{\pm}0.25kg$) where 67% of the silage dry matter was replaced with the UWCW and the lowest value for DMIF was observed in the control treatment ($5.03{\pm}0.23kg$). The DLWG did not differ significantly between the treatments. However, treatment "S100" showed a trend towards a superior DLWG. Feed conversion ratio in the control treatment differed significantly from "S67" and "S33". The addition of the UWCW in the animals' diet resulted in the lower FCR There was no effect of type of energy supplement on any aspect of performance either overall or in interaction with grass silage: UWCWS ratio. The regression and correlation coefficients for DMIF (r = 5.22 + 0.0184x*), DLWG (r = $1.04-0.00086x^{NS}$) and FCR (r = 4.78 = 0.022x*) on the inclusion of UWCW in the diet were calculated. The effect of the inclusion of UWCW on the overall digestibility coefficients was significant (p < 0.05). The addition of the UWCWS in the diet decreased the digestibility of the DM, OM, ADF and NFE but effect on the protein digestibility was non significant. The results of present study suggests that a DLWG slightly over 1 kg can be achieved with UWCW during the store period (period in which animal performance targets are low especially during winter) and the prediction of ME was overestimated as the high intake of DM did not reflect in improved animal performance.

• Clean Technology
• /
• v.26 no.3
• /
• pp.177-185
• /
• 2020

All coal ash, generated from coal-fired power plants, is entirely dumped onto a landfill site. As coal ash contains 80% fly ash, a clean floating process was developed in this study to recover useful components from coal ash and to use them as high value-added industrial materials. When the unburned carbon (UC) was recovered from the fly ash, soybean oil, an eco-friendly vegetable oil, was used as collector instead of a non-ionic kerosene collector to prevent the occurrence of odor from the kerosene. After the UC was separated by flotation, particulate ceramic microsphere (CM) was recovered, without generating acidic wastewater, through hydro-cyclone instead of sulfuric acid solution in order to separate ceramic microsphere (CM) and cleaned ash (CA) from the residue. By utilizing soybean oil as a collector, the recovery rate of UC turned high at 85.8% due to the increased adsorption of UC, the high viscosity of soybean oil, and the increase in floating properties caused by the linoleic acid contained in soybean oil. All of the combustible components contained in the recovered UC were carbon components, with the carbon content registering high when soybean oil was used. The recovered UC had many pores with a rough surface; thus, it could be easily ground and then used as an industrial material for its fine particles. The CM and CA recovered by the clean separation process using hydro-cyclone had a spherical shape, and the particles were clearly separated without clumping together. The average diameter (D₅₀) of the particles was 5 ㎛, so it was possible to realize the atomization of CM through a process change.

• Journal of Distribution Science
• /
• v.13 no.9
• /
• pp.95-101
• /
• 2015

Purpose - This study provides evidence of the impact of the mandatory adoption of Korean equivalents to International Financial Reporting Standards (K-IFRS) on accounting quality. K-IFRS uses fair value as a basis of measurement and is characterized by principle-based standards. These characteristics can lead to a decrease in conservatism. Therefore, this study aims to examine whether or not there is a change in the level of conservatism before and after the enforcement of K-IFRS (2007~2014). By comparing 2007 through 2008 and 2013 through 2014 (excluding 2009 to 2012), we test "the temporary adjustment phenomenon" and document an overall decline in the degree of conservatism after the adoption of K-IFRS. Research design, data, and methodology - Our sample is comprised of data of all listed Korea Composite Stock Price Index (KOSPI) manufacturing distribution companies in Korea from 2007 to 2014, which yields the pooled sample of 4,412 (panel A) and 1,915 (panel B) firm-year observations for hypotheses 1 and 2. In line with recent literature, we adopt the Givoly and Hayn (2000) model, which recomputes the non-operating accruals, excluding two components that are most likely to capture the effect of restructuring activities: special items and gains or losses from discontinued operations. In addition, we also use these variables: SIZE, LEV, INV_CYCLE, ROA, OWN, and FOR. Results - Our sample period spans 2007 to 2014. This offers evidence on the effect of the mandatory adoption of IFRS on conservatism. Our findings can be summarized as follows. First, in panel A, for mandatory K-IFRS adoption (2011), we do not find any significant evidence of conservatism. We can guess that the "temporary adjustment phenomenon" is the reason that we do not find significant evidence of conservatism. Second, we investigate panel B from 2009 to 2012. We document an overall decline in the degree of conservatism after the adoption of K-IFRS. We can assume that these results are due to "the temporary adjustment phenomenon." Conclusions - This study finds that conservatism significantly decreased after IFRS adoption. In particular, this study makes the initial effort to elucidate "the temporary adjustment phenomenon" to analyze the effect of K-IFRS on conservative accounting. We argue that K-IFRS are conceptually conservative but that inappropriate application of the conservatism principles is likely to prevent financial reporting from reaching the level of conservatism targeted by the IASB. Overall, this paper contributes to the literature on IFRS and can be useful to capital market supervisors who are monitoring the trends of the firms implementing K-IFRS. Additionally, our results inform stakeholders of the potentially negative effect of the greater flexibility permitted by IFRS and/or lack of appropriate enforcement on key dimensions of accounting quality. This has important implications for Korean regulators and standard setters as they review the cost and benefits of IFRS. Our study also sheds light on the importance of the institutional environment in achieving the targeted objectives for improving financial reporting quality.

• Food Science of Animal Resources
• /
• v.37 no.5
• /
• pp.663-669
• /
• 2017

This study were estimated the contribution of carcass traits to unit price, to analyze the marbling score as a categorical variable rather than a numerical variable, and to develop an optimal model that also includes the holiday effect and the raising period. The data for this study were acquired from the Quality Evaluation of the Korea Institute for Animal Products, and consisted of the trading records of 1,613,699 heads at 12 wholesale markets from 2010 to 2014. The unit price of a cow was estimated from the following parameters: -52.50 Won/mm, $8.93Won/cm^2$, 7.20 Won/kg, and -1.04 Won/day for backfat thickness, eye muscle area, carcass weight, and raising period, respectively. Parameters for the dummy variables of marbling scores varied from 0 to 8328.74 Won/kg, which means that each marbling score grade had a different price value. The unit price of a steer was estimated from the following parameters: - 92.12 Won/mm, $20.22Won/cm^2$, 1.30 Won/kg, and -1.72 Won/day for backfat thickness, eye muscle area, carcass weights, and raising period, respectively. Parameters for dummy variables of marbling scores varied from 0 to 7338.80 Won/kg, which means that the grades of each marbling score had different price values. The unit price of sales during traditional holidays was significantly higher (827.71 Won/kg for cows, and 645.15 Won/kg for steers) than during non-holidays.We conclude that the use of categorical values for marbling scores would be needed to evaluate the price of Hanwoo beef using multiple regression analysis based on carcass traits and environmental factors.

• Journal of the Korean GEO-environmental Society
• /
• v.11 no.1
• /
• pp.19-26
• /
• 2010

The batch experiments and response surface methodology (RSM) have been applied to the investigation of the cadmium (Cd) adsorption by coal fly ash (CFA). CFA having maximum Cd removal mass of 8.51 mg/g were calculated from Langmuir model. Cd removal reaction with different initial pH ranged from 4 to 9. When the initial pH was higher, Cd was removed more by adsorption and precipitation. These results suggest that the lower pH cause an increase of $H^+$ ion concentration which competed with Cd ions for exchange sites in CFA. Also, The Cd adsorption was mathematically described as a function of parameters initial Cd concentration ($X_1$), initial pH ($X_2$), and initial CFA mass ($X_3$) being modeled by use of the Box-Behnken methods. Empirical models were developed to describe relationship between the experimental variables and response. Statistical analysis indicates that tree factors ($X_1$, $X_2$, and $X_3$) on the linear term (main effects), and tree factors ($X_1X_2$, $X_1X_3$, and $X_2X_3$) on the non-linear term (Interaction effect; cross-product) had significant effects, respectively. In this case, the value of the adjusted determination coefficient (adjusted $R^2=0.9280$) was closed to 1, showing a high significance of the model. Statistical results showed the order of Cd removal at experimental factors to be initial initial pH > initial Cd concentration > initial CFA mass.

• Journal of the Korean Geotechnical Society
• /
• v.17 no.6
• /
• pp.15-24
• /
• 2001

In order to evaluate undrained shear strength of clayey soils using Piezocone Penetration Test (CPTu), piezoncone factor is utilized. Commonly, piezoncone factors determined by empirical basis were preferred, which were established by correlation between measurements of piezocone test and undrained strengths obtained from other shearing tests. However, previous studies on the empirical piezocone factors were site-specific and there have been no systematic investigations on the effect of both engineering characteristics of clayey soils and soil non-homogeneity on the piezocone factor. Accordingly, the direct application of the previous results to Korean clayey soils without verification may be inappropriate. In this study, empirical piezocone factors are evaluated by comparing 46 CPTu results of 10 test sites with undrained shear strength obtained from Field Vane Test (FVT) and laboratory triaxial tests. Their reliabilities are investigated by the comparison with the previous piezocone factors and the deviation of data distribution from the mean values. And the effects of referencing test methods and typical engineering characteristics of clayey soils such as overconsolidation ratio (OCR) and plastic Index (I$_{p}$) are examined. Because piezocone factors obtained for various soil conditions are widely distributed, it is not appropriate to use the mean value as a representative. Instead, it is recommended to apply the piezocone factors with OCR, which is found to be a major factor in deriving piezocone factor. The necessitated piezocone factors are presented.d.

• Korean Journal of Health Education and Promotion
• /
• v.6 no.1
• /
• pp.22-33
• /
• 1989

Rapid socio-cultural and economic changes in the country has brought with it changes in the society's value system. For a traditional society that is increasingly being exposed to modernization but where sex norms are still very restrictive, the adolescent sexual mores takes on added significance. Adolescents are caught between two opposing forces, the changing environment that allows for freer and liberal mores and the traditional society that cannot keep pace with the changing environment and therefore demands resistance to changes. This paper focuses on problems of adolescent sexuality in this country and considers the countermeasures for the existing problems. Amongst the problems are: (a) increasingly younger age of the adolescents who start sexual intercourse (b) non-use of contraception, (c) unwanted pregnancies, (d) increase in the number of induced abortion and (e) increase in the number of unwanted children and unmarried mothers. The Korean adolescent's sexual behavior seems to follow that of the developed countries. In other words, many western modes of life and sexual values seem to bave been copied in Korea and yet Korean adolescents lack in their knowledge of sex related matters such as reproductive physiology and contraception. Among middle and high school students, female students are reported to have less knowledge on sex than male students according to a 1988 survey by KIPH. Even among the unmarried famale factory workers, only 42.5 percent replied they know of the condom, and 25.1 percent and 23.1 percent said they had knowledge of spermicide and menstrual regulation respectively. However, 14.9 percent and 13.9 percent reported that they had a knowledge of the loop and female sterilization respectively according to the 1984 study by KIPH. Among the middle school students 0.8 percent said they had experience in sexual intercourse, while 7.3 percent of the high school students reported having had sexual intercourse. The sexual intercourse experience rate among the unmarried female factory workers is 37.8 percent. Among those female factory workers with sexual experience, 46.7 percent had more than one sex partners. Only 39.1 percent of male students and 18.9 percent of female students among those with sexual intercourse experience have used contraceptives. mostly condoms and oral pills 45.1 percent of female factory workers with sexual intercourse experience used contraceptives such as pills, condoms and rhythm methods. The pregnancy experience rate among the female factory workers who had experience in sexual intercourse is 29.5 percent, which is 11.1 percent among the total respondents. Out of the 102 pregnant female workers, 98 workers(96.1 percent) terminated their pregnancy by induced abortion and 2 workders(2 percent) in natural abortion, while 1 worker(1 percent) was in pregnancy and another 1 worker had normal birth that was subsequently sent to orphanage. In order to cope with the problem of adolescent sexuality, a drastic and strong policy measures should be taken by the government. The most effective countermeasure to the adolescent sexual problems appears to the education. The sex and population education in the school is very much in need. In addition, sex education program through mass media and at the job sit-should be promoted for a healthy development of adolescents' sexual behavior. Also, the existing national family planning program, which has focused on the married couples, should be extended to the unmarried people in its scope and contents of the program.