• Journal of Intelligence and Information Systems
• /
• v.24 no.2
• /
• pp.221-241
• /
• 2018

Deep learning is getting attention recently. The deep learning technique which had been applied in competitions of the International Conference on Image Recognition Technology(ILSVR) and AlphaGo is Convolution Neural Network(CNN). CNN is characterized in that the input image is divided into small sections to recognize the partial features and combine them to recognize as a whole. Deep learning technologies are expected to bring a lot of changes in our lives, but until now, its applications have been limited to image recognition and natural language processing. The use of deep learning techniques for business problems is still an early research stage. If their performance is proved, they can be applied to traditional business problems such as future marketing response prediction, fraud transaction detection, bankruptcy prediction, and so on. So, it is a very meaningful experiment to diagnose the possibility of solving business problems using deep learning technologies based on the case of online shopping companies which have big data, are relatively easy to identify customer behavior and has high utilization values. Especially, in online shopping companies, the competition environment is rapidly changing and becoming more intense. Therefore, analysis of customer behavior for maximizing profit is becoming more and more important for online shopping companies. In this study, we propose 'CNN model of Heterogeneous Information Integration' using CNN as a way to improve the predictive power of customer behavior in online shopping enterprises. In order to propose a model that optimizes the performance, which is a model that learns from the convolution neural network of the multi-layer perceptron structure by combining structured and unstructured information, this model uses 'heterogeneous information integration', 'unstructured information vector conversion', 'multi-layer perceptron design', and evaluate the performance of each architecture, and confirm the proposed model based on the results. In addition, the target variables for predicting customer behavior are defined as six binary classification problems: re-purchaser, churn, frequent shopper, frequent refund shopper, high amount shopper, high discount shopper. In order to verify the usefulness of the proposed model, we conducted experiments using actual data of domestic specific online shopping company. This experiment uses actual transactions, customers, and VOC data of specific online shopping company in Korea. Data extraction criteria are defined for 47,947 customers who registered at least one VOC in January 2011 (1 month). The customer profiles of these customers, as well as a total of 19 months of trading data from September 2010 to March 2012, and VOCs posted for a month are used. The experiment of this study is divided into two stages. In the first step, we evaluate three architectures that affect the performance of the proposed model and select optimal parameters. We evaluate the performance with the proposed model. Experimental results show that the proposed model, which combines both structured and unstructured information, is superior compared to NBC(Naïve Bayes classification), SVM(Support vector machine), and ANN(Artificial neural network). Therefore, it is significant that the use of unstructured information contributes to predict customer behavior, and that CNN can be applied to solve business problems as well as image recognition and natural language processing problems. It can be confirmed through experiments that CNN is more effective in understanding and interpreting the meaning of context in text VOC data. And it is significant that the empirical research based on the actual data of the e-commerce company can extract very meaningful information from the VOC data written in the text format directly by the customer in the prediction of the customer behavior. Finally, through various experiments, it is possible to say that the proposed model provides useful information for the future research related to the parameter selection and its performance.

• Proceedings of the Korea Hospitality Industry Research Society Conference
• /
• 2002.11a
• /
• pp.83-98
• /
• 2002

The old testament of the Bible has written the milk and curd. God said, I will ive you to how the milk and honey. The present study was designed to examine the effects of different homogenization pressure, homogenization temperature and $\beta$-cyclodextrin concentration on cholesterol removal rate of cheese, and to optimize the factors of cheese manufacture Process. In addition, the characteristics from cholesterol removed cheese and control are compared in the rheological and ensory analysis. The optimized process condition for cholesterol removal was for homogenization pressure, 74$^{\circ}C$ for homogenization temperature and 2% for $\beta$-cyclodextrin concentration, it showed 875% of the highest cholesterol removal rate in milk. Therefore, manufacture conditions of cholesterol removed cheese were chosen 74$^{\circ}C$ for homogenization temperature, for homogenization pressure, and I or 2% for $\beta$-cyclodextrin concentration. Cholesterol removed cheese and control were compared with yield, cholesterol removal, meltability, stretchability, textural properties and sensory analysis. Cholesterol content of control cheese containing 23.8% milk fat was cheese made from milk treated with 2% $\beta$-cyclodextrin and homogenization pressure was cholesterol removal. Yield of cholesterol removed cheese. As the homogenization pressure increased, oiling off reduced with showed better surface appearance. Stretchability of cholesterol removed cheese was lower 5~10cm than over 30cm of control. Meltability of cholesterol removed cheese also was lower than control. The hardness, gumminess, chewiness reduced to respectively. In the result of sensory analysis, treatment of homogenization for cholesterol removed cheese improved appearance and flavor, however texture fell. In addition, the resent result of the study indicated that about 75% of cholesterol in cheese could be removed, and the possibility of development of cholesterol removed cheese was observed. We have hope to research manufacture cheese global wide.

• The Korean Journal of Nuclear Medicine
• /
• v.38 no.3
• /
• pp.259-267
• /
• 2004

Purpose: NEMA NU2-2001 was proposed as a new standard for performance evaluation of whole body PET scanners. in this study, system performance of Siemens CTI ECAT EXACT 47 PET scanner including spatial resolution, sensitivity, scatter fraction, and count rate performance in 2D and 3D mode was evaluated using this new standard method. Methods: ECAT EXACT 47 is a BGO crystal based PET scanner and covers an axial field of view (FOV) of 16.2 cm. Retractable septa allow 2D and 3D data acquisition. All the PET data were acquired according to the NEMA NU2-2001 protocols (coincidence window: 12 ns, energy window: $250{\sim}650$ keV). For the spatial resolution measurement, F-18 point source was placed at the center of the axial FOV((a) x=0, and y=1, (b)x=0, and y=10, (c)x=70, and y=0cm) and a position one fourth of the axial FOV from the center ((a) x=0, and y=1, (b)x=0, and y=10, (c)x=10, and y=0cm). In this case, x and y are transaxial horizontal and vertical, and z is the scanner's axial direction. Images were reconstructed using FBP with ramp filter without any post processing. To measure the system sensitivity, NEMA sensitivity phantom filled with F-18 solution and surrounded by $1{\sim}5$ aluminum sleeves were scanned at the center of transaxial FOV and 10 cm offset from the center. Attenuation free values of sensitivity wire estimated by extrapolating data to the zero wall thickness. NEMA scatter phantom with length of 70 cm was filled with F-18 or C-11solution (2D: 2,900 MBq, 3D: 407 MBq), and coincidence count rates wire measured for 7 half-lives to obtain noise equivalent count rate (MECR) and scatter fraction. We confirmed that dead time loss of the last flame were below 1%. Scatter fraction was estimated by averaging the true to background (staffer+random) ratios of last 3 frames in which the fractions of random rate art negligibly small. Results: Axial and transverse resolutions at 1cm offset from the center were 0.62 and 0.66 cm (FBP in 2D and 3D), and 0.67 and 0.69 cm (FBP in 2D and 3D). Axial, transverse radial, and transverse tangential resolutions at 10cm offset from the center were 0.72 and 0.68 cm (FBP in 2D and 3D), 0.63 and 0.66 cm (FBP in 2D and 3D), and 0.72 and 0.66 cm (FBP in 2D and 3D). Sensitivity values were 708.6 (2D), 2931.3 (3D) counts/sec/MBq at the center and 728.7 (2D, 3398.2 (3D) counts/sec/MBq at 10 cm offset from the center. Scatter fractions were 0.19 (2D) and 0.49 (3D). Peak true count rate and NECR were 64.0 kcps at 40.1 kBq/mL and 49.6 kcps at 40.1 kBq/mL in 2D and 53.7 kcps at 4.76 kBq/mL and 26.4 kcps at 4.47 kBq/mL in 3D. Conclusion: Information about the performance of CTI ECAT EXACT 47 PET scanner reported in this study will be useful for the quantitative analysis of data and determination of optimal image acquisition protocols using this widely used scanner for clinical and research purposes.

• Korean Journal of Agricultural Science
• /
• v.11 no.1
• /
• pp.146-159
• /
• 1984

This study was the contrast of the compressive strength, the tensile strength, the reducing ratio and the flow of mortar using dispersing agent and high fluid admix. 1. The admix ratio of chemical admixtures espressing maximum strength appeared the same result high fluid admix SP was 0.6%, the dispersing agents LG and C211 were 0.2%, SK was 0.3%, C376 was 0.5%. But two or three times more than standard quantity made the strength's fast lowness, which influenced bad to wateriness and retard the soli-dification. 2. When proper quantity of chemical admixture was used, the increment of compressive strength was as follows. High fluid admix SP was 40.7% and the average increasing rate of dispersing agents(C211 was 19.5%, LG was 19.1%, C376 was 17.9%) was 18.7% more than normal mortar in the codition of 7 days. Also, in the condition of 28 days, high fluid admix SP was about 24.4% and the average of dispersing agents(LG was 21.1%, C211 was 16.4%, SK was 11.1%, C376 was 7.6%) was 14.1%. 3. When proper quantity of chemical admixture was used, the increment of tensile strength was as follows. High fluid admixture SP was 26.6% and the average increasing agents(SK was 16.0%, C376 was 14.7%, LG was 10%, C211 was 5.8%) was 11.6%. Also, in the condition of 28 days, high fluid admix SP was 16.5% and the average increasing rate of dispersing agents(LG was 19.1%, SK was 10.6%, C211 was 10.1%, C376 was 8.7%) was 12.1%. 4. As for the reducing ratio of each dispersing agent, he flow of mortar was less than the slump of concrete. That is; the reducing ratio of concrete was 15% adding each dispersing agent, but the reducing ratio of mortar was in the range of from 5.8% to 13.5% in 1 : 1 mixture, from 7.6% to 14.2% in 1 : 2, from 9.5% to 18.8% in 1 : 3. 5. The fluidity of each chemical admixture was as follows. High fluid admix SP in the condition of 1: 1 and 1 : 2 showed the best result than other dispersing agent and 1 : 3 showed the same result like other agents. Therefore these good dispersing agents were suitable in the prepact concrete construction using intrusion mortar.

• Journal of Embryo Transfer
• /
• v.22 no.1
• /
• pp.53-61
• /
• 2007

This study was investigated factors affecting the pregnancy rates after transfer of pronuclear microinjected embryos for the production of transgenic Korean black goats. Embryo transfer was carried out in 343 recipient Korean black goats from September 1999 to June 2000. Estrus was induced by the insertion of intravaginal progesterone devices $CIDR^(R)$ for 2 weeks. A single injection of 400 IU equine chorionic gonadotropin was administered at 48h before $CIDR^(R)$ removal to increase the proportion of does cycling and ovulation rate. Good quality embryos were prepared by microinjection of DNA into the pronuclei of fertilized goat oocyte and cultured in vitro. Pronuclear microinjected $1{\sim}8$ cell stage embryos were surgically transferred into the oviducts of the recipient at day 4 or 5 following $CIDR^(R)$ removal, and morula to blastocyst stage embryos were surgically transferred into uterus at day 9. Pregnancy was diagnosed by transrectal ultrasound scanning at $20{\sim}30d$ and 8 weeks following embryo transfer. The pregnancy rate was affected by several factors, such as estrus induction, the number of previous transfer, transfer site, stage of CL (corpus luteum), the number of recipient CL, stage of embryos and the number of transferred embryo. The pregnancy rate was significantly higher in recipients that came into estrus naturally than recipients that induced to come into estrus with $CIDR^(R)$(59.1% vs. 36.8%; P<0.05). The pregnancy rate was higher when the embryos were transferred into the left oviduct than transferred into the right oviduct (42.9% vs. 35.3%; P<0.05). The pregnancy rate of recipients with $CH_1$ (early) stage corpus hemorrhagicum in ovary was hi틴or than recipient with $CH_3$ (late) stage hemorrhagicum (47.5% vs. 17.9%; P<0.01). Higher pregnancy rates were obtained by transfer of 1-cell stage embryos into oviduct while late blastocysts (51.6% vs. 66.7%; P<0.01) into uterus. The pregnancy rates when 3 embryos were transferred to recipients were significantly higher than when 2 embryos we.e transferred (47.6% vs. 27.0%; P<0.05). Although there were no significant difference among the group, adhesion of reproductive organs, uterine size, ovulation rate of recipients, presence of large follicle and difficulty of transfer affected pregnancy rate of recipient. Higher pregnancy rates were obtained in the recipients with $8{\sim}15m$ diameter uterine horn as compared to the recipients with <5m diameter or >20mm diameter uterine hem (38.9%, 20% vs. 18.2%), in the recipients with large follicle in the ovulated ovary ipsilaterally (53.6% vs. 37.1%) and in the transfer which was carried out easily (39.2% vs. 27.8%, 0%). In conclusion, the high rate of pregnancy was achieved following transfer of pronuclear microinjected embryos when three or four 1-cell stage embryos were transferred into oviduct with $CH_1$ stage corpus hemorrhagicum in the ovary of recipient which came into estrus naturally.

• Journal of Intelligence and Information Systems
• /
• v.23 no.4
• /
• pp.147-168
• /
• 2017

There have been many studies on accurate stock market forecasting in academia for a long time, and now there are also various forecasting models using various techniques. Recently, many attempts have been made to predict the stock index using various machine learning methods including Deep Learning. Although the fundamental analysis and the technical analysis method are used for the analysis of the traditional stock investment transaction, the technical analysis method is more useful for the application of the short-term transaction prediction or statistical and mathematical techniques. Most of the studies that have been conducted using these technical indicators have studied the model of predicting stock prices by binary classification - rising or falling - of stock market fluctuations in the future market (usually next trading day). However, it is also true that this binary classification has many unfavorable aspects in predicting trends, identifying trading signals, or signaling portfolio rebalancing. In this study, we try to predict the stock index by expanding the stock index trend (upward trend, boxed, downward trend) to the multiple classification system in the existing binary index method. In order to solve this multi-classification problem, a technique such as Multinomial Logistic Regression Analysis (MLOGIT), Multiple Discriminant Analysis (MDA) or Artificial Neural Networks (ANN) we propose an optimization model using Genetic Algorithm as a wrapper for improving the performance of this model using Multi-classification Support Vector Machines (MSVM), which has proved to be superior in prediction performance. In particular, the proposed model named GA-MSVM is designed to maximize model performance by optimizing not only the kernel function parameters of MSVM, but also the optimal selection of input variables (feature selection) as well as instance selection. In order to verify the performance of the proposed model, we applied the proposed method to the real data. The results show that the proposed method is more effective than the conventional multivariate SVM, which has been known to show the best prediction performance up to now, as well as existing artificial intelligence / data mining techniques such as MDA, MLOGIT, CBR, and it is confirmed that the prediction performance is better than this. Especially, it has been confirmed that the 'instance selection' plays a very important role in predicting the stock index trend, and it is confirmed that the improvement effect of the model is more important than other factors. To verify the usefulness of GA-MSVM, we applied it to Korea's real KOSPI200 stock index trend forecast. Our research is primarily aimed at predicting trend segments to capture signal acquisition or short-term trend transition points. The experimental data set includes technical indicators such as the price and volatility index (2004 ~ 2017) and macroeconomic data (interest rate, exchange rate, S&P 500, etc.) of KOSPI200 stock index in Korea. Using a variety of statistical methods including one-way ANOVA and stepwise MDA, 15 indicators were selected as candidate independent variables. The dependent variable, trend classification, was classified into three states: 1 (upward trend), 0 (boxed), and -1 (downward trend). 70% of the total data for each class was used for training and the remaining 30% was used for verifying. To verify the performance of the proposed model, several comparative model experiments such as MDA, MLOGIT, CBR, ANN and MSVM were conducted. MSVM has adopted the One-Against-One (OAO) approach, which is known as the most accurate approach among the various MSVM approaches. Although there are some limitations, the final experimental results demonstrate that the proposed model, GA-MSVM, performs at a significantly higher level than all comparative models.

• Radiation Oncology Journal
• /
• v.19 no.1
• /
• pp.53-65
• /
• 2001

Purpose : To improve the local control of patients with nasopharyngeal cancer, we have implemented 3-D conformal radiotherapy and forward intensity modulated radiation therapy (IMRT) to used of compensating filters. Three dimension conformal radiotherapy with intensity modulation is a new modality for cancer treatments. We designed 3-D treatment planning with 3-D RTP (radiation treatment planning system) and evaluation dose distribution with tumor control probability (TCP) and normal tissue complication probability (NTCP). Material and Methods : We have developed a treatment plan consisting four intensity modulated photon fields that are delivered through the compensating tilters and block transmission for critical organs. We get a full size CT imaging including head and neck as 3 mm slices, and delineating PTV (planning target volume) and surrounding critical organs, and reconstructed 3D imaging on the computer windows. In the planning stage, the planner specifies the number of beams and their directions including non-coplanar, and the prescribed doses for the target volume and the permissible dose of normal organs and the overlap regions. We designed compensating filter according to tissue deficit and PTV volume shape also dose weighting for each field to obtain adequate dose distribution, and shielding blocks weighting for transmission. Therapeutic gains were evaluated by numerical equation of tumor control probability and normal tissue complication probability. The TCP and NTCP by DVH (dose volume histogram) were compared with the 3-D conformal radiotherapy and forward intensity modulated conformal radiotherapy by compensator and blocks weighting. Optimization for the weight distribution was peformed iteration with initial guess weight or the even weight distribution. The TCP and NTCP by DVH were compared with the 3-D conformal radiotherapy and intensitiy modulated conformal radiotherapy by compensator and blocks weighting. Results : Using a four field IMRT plan, we have customized dose distribution to conform and deliver sufficient dose to the PTV. In addition, in the overlap regions between the PTV and the normal organs (spinal cord, salivary grand, pituitary, optic nerves), the dose is kept within the tolerance of the respective organs. We evaluated to obtain sufficient TCP value and acceptable NTCP using compensating filters. Quality assurance checks show acceptable agreement between the planned and the implemented MLC(multi-leaf collimator). Conclusion : IMRT provides a powerful and efficient solution for complex planning problems where the surrounding normal tissues place severe constraints on the prescription dose. The intensity modulated fields can be efficaciously and accurately delivered using compensating filters.

• Radiation Oncology Journal
• /
• v.20 no.4
• /
• pp.323-327
• /
• 2002

Purpose : To evaluate the outcome of early stage non-small cell lung cancer patients who were treated with radiation therapy alone and define the optimal radiotherapeutic regimen for these patients. Materials and Methods : A retrospective review was peformed on patients with sage I or II non-small cell carcinoma of the lung that were treated at our institution between June, 1987 and May, 2000. A total of 21 patients treated definitively with radiation therapy alone were included in this study. The age of the patients ranged from 53 to 81 years with a median of 66 years. All the patients were male. The medical reasons for inoperability were lack of pulmonary reserve, cardiovascular disease, poor performance status, old age, and patient refusal in the decreasing order. Pathological evidence was not adequate to characterize the non-small cell subtype in two patients. Of the remaining 19 patients, 16 had squamous cell carcinoma and 3 had adenocarcinoma. Treatment was given with conventional fractionation, once a day, five times a week. The doses to the primary site ranged from 56 Gy to 59 Gy. No patients were lost to follow-up. Results : The overall survival rates for the entire group at 2, 3 and 5 years were 41, 30 and $21\%$, respectively. The cause specific survivals at 2, 3 and 5 years were 55, 36 and $25\%$, respectively. An intercurrent disease was the cause of death in two patients. The cumulative local failure rate at 5 years was $43\%$. Nine of the 21 patients had treatment failures after the curative radiotherapy was attempted. Local recurrences as the first site of failure were documented in 7 patients. Therefore, local failure alone represented $78\%$ of the total failures. Those patients whose tumor sizes were less than 4 cm had a significantly better 5 year disease free survival than those with tumors greater than 4 cm $(0\%\;vs\;36\%)$. Those patients with a Karnofsky performance status less than 70 did not differ significantly with respect to actuarial survival when compared to those with a status greater than 70 $(25\%\;vs\;26\%,\;p>0.05)$. Conclusion : Radiation therapy 리one is an effective and safe treatment for early stage non-small ceil lung cancer patients who are medically inoperable or refuse surgery. Also we believe that a higher radiation dose to the primary site could improve the local control rate, and ultimately the overall survival rate.

• Applied Biological Chemistry
• /
• v.18 no.1
• /
• pp.16-22
• /
• 1975

This experiment was carried out to investigate the physiological characteristics of two original yeasts, 5-Y-5 and 6-Y-6, which selected from 24 Takju yeasts and three mutants, 30-24,30-81 and 40-27. induced from two original yeasts by the irradiation of UV light. The results were summarized as follows. 1) Alcohol tolerances of three mutants were decreased in some degree as compared with those of original yeasts. 2) Tolerances of lactic and citric acids of acid producing mutant 30-81, was increased than those of original yeasts. 3) In the case of using ammonium sulfate as a nitrogen source, two original yeasts and three mutants required Ca-pantothenate as a essential growth factor and four strains of yeasts except the mutant, 30-81, required biotin as a stimulated growth factor, When asparagine was used as a nitrogen source, two original yeasts and three mutants showed the same as above result but the stimulated effect of biotin was far less. 4) Propagation powers of the mutants were weaken than those of original yeasts, particular that of acid producing mutant, 30-81, was the weakest in the three mutants. 5) The optimum temperature for fermentation of original yeasts were $30^{\circ}C\;to\;35^{\circ}C$ but three mutants were $25^{\circ}C\;to\;30^{\circ}C$. 6) The optimum pH for fermentation of original yeasts were pH 5 to 6, and there is no appreciable difference between original yeasts and three mutants. The fermentation power of mutant,30-81, was decreased more rapidly than those of other mutants according to approach neutral. Three mutants were more sensible to heat than original yeasts. 7) Two original yeasts and three mutants were inhibited more over 20 percent of sugar for fermentation and three mutants were more sensible to sugar concentration than original yeasts.

• Korean Journal of Fisheries and Aquatic Sciences
• /
• v.13 no.2
• /
• pp.65-80
• /
• 1980

Processing conditions of the krill products such as paste food, krill protein concentrate, seasoned dried krill, powdered seasoning, meat ball, and snack have been examined and the quality was evaluated chemically and organoleptically. In the processing of paste food, krill juice was yielded $71\%$ and krill scrap $29\%$. The yields of paste and broth from the krill juice showed $53\%$ and $43\%$, respectively. In amino acid composition of the krill paste, proline, glutamic acid, aspartic acid, lysine, and leucine were abundant, while histidine, methionine, tyrosine, serine and threonine were poor. The optimum condition for solvent extraction in the processing of krill protein concentrate was the 5 times repetitive extraction using isopropyl alcohol at $80^{\circ}C$ for 5 mins. The yield of krill protein concentrate when used fresh frozen materials was $10.2\%$ in isopropyl alcohol solvent and $8.8\% in ethyl alcohol, and when used preboiled frozen materials, the yield was $13.0\%$ in isopropyl alcohol and $11.8\%$ in ethyl alcohol. Amino acid composition of krill protein concentrate showed a resemblance to that of fresh frozen krill meat. In quality comparison of the seasoned dried krill, hot air dried krill was excellent as raw materials and sun dried krill was slightly inferior to hot air dried krill, but preboiled frozen krill showed the poorest quality. The result of quality evaluation for seasoning made by combination of dried powdered krill, parched powdered sesame, salt, powdered beef extract, monosodium glutamate, powdered red pepper and ground pepper showed that the hot air dried krill was good in color and sundried krill was favorable in flavor. When krill meat ball was prepared using wheat flour, monosodium glutamate and salt as side materials, the quality of the products added up to $52\%$ of krill meat was good and the difference in quality upon the results of the organoleptic test for raw materials was not recognizable between fresh frozen and preboiled frozen krill. In the experiment for determining the proper amount of materials such as dried Powdered krill, $\alpha-starch$, sweet potato starch, sugar, salt, monosodium glutamate, glycine, potassium tartarate, ammonium bicarbonate, and sodium bicarbonate in processing krill snack, sample B(containing $7.7\%$ of dried powdered krill) and sampleC (containing $10.8\%$ of dried powdered krill) showed the most palatable taste from the view point of organoleptic test. Sweet potato starch in testing side materials was good in the comparison of suitability for processing krill snack. Corn starch and kudzu starch were slightly inferior to sweet potato starch, while wheat flour was not proper for processing the snack. In the experiment on frying method, oil frying showed better effect than salt frying and the suitable range of frying temperature was $210-215^{\circ}C$.