• Journal of Intelligence and Information Systems
• /
• v.18 no.4
• /
• pp.43-57
• /
• 2012

To enhance the competitive advantage in a constantly changing business environment, an enterprise management must make the right decision in many business activities based on both internal and external information. Thus, providing accurate information plays a prominent role in management's decision making. Intuitively, historical data can provide a feasible estimate through the forecasting models. Therefore, if the service department can estimate the service quantity for the next period, the service department can then effectively control the inventory of service related resources such as human, parts, and other facilities. In addition, the production department can make load map for improving its product quality. Therefore, obtaining an accurate service forecast most likely appears to be critical to manufacturing companies. Numerous investigations addressing this problem have generally employed statistical methods, such as regression or autoregressive and moving average simulation. However, these methods are only efficient for data with are seasonal or cyclical. If the data are influenced by the special characteristics of product, they are not feasible. In our research, we propose a forecasting framework that predicts service demand of manufacturing organization by combining Case-based reasoning (CBR) and leveraging an unsupervised artificial neural network based clustering analysis (i.e., Self-Organizing Maps; SOM). We believe that this is one of the first attempts at applying unsupervised artificial neural network-based machine-learning techniques in the service forecasting domain. Our proposed approach has several appealing features : (1) We applied CBR and SOM in a new forecasting domain such as service demand forecasting. (2) We proposed our combined approach between CBR and SOM in order to overcome limitations of traditional statistical forecasting methods and We have developed a service forecasting tool based on the proposed approach using an unsupervised artificial neural network and Case-based reasoning. In this research, we conducted an empirical study on a real digital TV manufacturer (i.e., Company A). In addition, we have empirically evaluated the proposed approach and tool using real sales and service related data from digital TV manufacturer. In our empirical experiments, we intend to explore the performance of our proposed service forecasting framework when compared to the performances predicted by other two service forecasting methods; one is traditional CBR based forecasting model and the other is the existing service forecasting model used by Company A. We ran each service forecasting 144 times; each time, input data were randomly sampled for each service forecasting framework. To evaluate accuracy of forecasting results, we used Mean Absolute Percentage Error (MAPE) as primary performance measure in our experiments. We conducted one-way ANOVA test with the 144 measurements of MAPE for three different service forecasting approaches. For example, the F-ratio of MAPE for three different service forecasting approaches is 67.25 and the p-value is 0.000. This means that the difference between the MAPE of the three different service forecasting approaches is significant at the level of 0.000. Since there is a significant difference among the different service forecasting approaches, we conducted Tukey's HSD post hoc test to determine exactly which means of MAPE are significantly different from which other ones. In terms of MAPE, Tukey's HSD post hoc test grouped the three different service forecasting approaches into three different subsets in the following order: our proposed approach > traditional CBR-based service forecasting approach > the existing forecasting approach used by Company A. Consequently, our empirical experiments show that our proposed approach outperformed the traditional CBR based forecasting model and the existing service forecasting model used by Company A. The rest of this paper is organized as follows. Section 2 provides some research background information such as summary of CBR and SOM. Section 3 presents a hybrid service forecasting framework based on Case-based Reasoning and Self-Organizing Maps, while the empirical evaluation results are summarized in Section 4. Conclusion and future research directions are finally discussed in Section 5.

• Korean Journal of Soil Science and Fertilizer
• /
• v.16 no.1
• /
• pp.28-35
• /
• 1983

The object of this study was to investigate the distribution of Cd fractions in paddy soils in relation to some soil characteristics and to find out the relationships between soil Cd fractions and Cd content in brown rice. Thirty six soils and rice samples were collected from the paddy field adjacent to zinc mining sites at harvest time in 1981. Total Cd content of brown rice samples was analyzed. A sequential extraction procedure was used to fractionate Cd in soils into the designated forms of exchangeable, adsorbed, organically bonded, carbonate, sulfide, and residual Cd. The results obtained were as follows: 1. The distribution of Cd fractions in soil showed a wide difference depending on soil properties. As an average value it was observed that organically bonded Cd amounted 43.7%; residual Cd, 6.5%; and other fractions, 10-15%. 2. With higher soil pH, organically bonded and carbonate Cd fraction tended to be higher but exchangeable fraction lower. Other forms of Cd showed no difference with soil reaction. 3. Organically bonded fraction was positively correlated with soil organic matter content, while others except adsorbed fraction showed an adverse tendency. 4. The relation of fraction distribution to soil C E C was similar to the case of organic matter. 5. Cadmium content in brown rice showed significant possitive correlation with organically bonded Cd content (r = 0.655) and carbonate, Cd content of soil (r = 0.328) but there was no significant correlation with contents of other forms.

• The Journal of the Petrological Society of Korea
• /
• v.4 no.1
• /
• pp.61-87
• /
• 1995

On the basis of lithology, the Precambrian Hongjesa Granitic Gneiss can be locally zoned into granoblastic granitic gneiss, porphyroblastic granitic gneiss, migmatitic gneiss from its center to the marginal part. There are no distinct differences in mineral assemblages by lithologic zoning, but it partly shows the change of mineral assemblage in the adjacent with migmatitic gneiss, thus mineral assemblage can be subdivided into Zone I and Zone II. In terms of mineral compositions, the characteristics of Zone I are coexisting K-feldspar+muscovite+sillimanite. The characteristics of Zone II are (1) breakdown of muscovite, (2) coexisting garnetScordierite, (3) coexisting garnet+cordierite + orthoamphibole. The Buncheon Granitic Gneiss is mainly composed of augen gneiss. In the adjacent area with Honjesa Granitic Gneisses, Buncheon Granitic Gneiss has the mineral assemblage of sillimanite+biotite+K-feldspar+(kyanite). Kyanite occurs as relict grains in the Buncheon and Hongjesa Granitic Gneissess. Kyanite shows anhedral to subhedral form and coexists with sillimanite in only one of these samples. Garnet from a migmatitic gneiss (Zone 11) has relatively high $X_{Fe}$ value in core and rim. Garnet from a porphyroblastic granitic gneiss(Zone I) has relatively homogemeous core but compositionally-zoned rim. Biotites show various colour from greenish-brown, brown to reddish brown at maximum adsorption. Also, the Ti, and Mg content in biotites increases from Zone I to Zone II. The plagioclases shows the chemical composition of $Ab_{84}An_{16}$ -$Ab_{70}An_{30}$ (oligoclase) in Zone I and $Ab_{70}An_{30}$ -$Ab_{50}An_{50}$(andesine) in Zone 11. These variations indicate that the gneisses in the study area experienced a upperamphibolite facies. The presence of kyanite as relict grains indicates that the metamorphic rocks in this area exprienced a high-temperature/medium-pressure type metamorphism, followed by high-temperaturellow-pressure metamorphism. Metamorphic P-T conditions for each gneiss estimated from various geothermobarometers and phase equilibria are 698-$729^{\circ}C$/6.3-11.3 kbar in augen gneiss, 621-$667^{\circ}C$/1.0-5.4 kbar in migmatitic gneiss, and 602-$624^{\circ}C$/1.9-3.4 kbar in porphyroblastic granitic gneiss. These data suggest that the study area was subjected to a clockwise P-T path with isothermal decompression (dP/dT=about 60 bar/$^{\circ}C$).

• Sleep Medicine and Psychophysiology
• /
• v.20 no.2
• /
• pp.75-81
• /
• 2013

Objectives: There are a lot of studies that analyze the interaction between the emotion of disgust and the functional brain images using fMRI and PET. But studies using sLORETA (standardized low resolution brain electromagnetic tomography) almost do not exist. The aim of this research is to explore the relationship of the emotion of disgust and the cortical activation using sLORETA analysis. Methods: Forty five healthy young adults ($27.1{\pm}2.6$ years) participated in the study. While they were watching 4 neutral images and 4 disgusting images associated with mutilation selected from the international affective picture system (IAPS), participants' EEGs were taken for 30 seconds per one picture. Through these obtained EEG data, sLORETA analysis was performed to compare EEGs associated with neutral and negative images. Results: During looking for visual disgusting stimulus, all participants reported unpleasantness, arousal and stress. In sLORETA analysis, the decrease of current density in theta wave was shown at left frontal superior gyrus (BA10) and middle gyrus (BA10, 11). This voxel cluster consists of a total of 11 voxels and the threshold of t value indicating statistically significant decreases in the current density (p<0.05) was -1.984. There were no differences between male and female in the degree of being disgusted by the stimuli. Conclusion: This finding may suggest that the activation of dorsolateral prefrontal cortex might be associated with regulating disgust emotion.

• Radiation Oncology Journal
• /
• v.22 no.2
• /
• pp.98-105
• /
• 2004

Purpose : To evaluate whether positron omission tomography (PET) with 2-[F-18]fluoro-2-deoxy-D-giucose(FDG) can be used to predict of early response to definitive aim radlotherapy (RT) in squamous cell carcinoma of the head and neck using response rate and locoreglonal control as study endpoints. Materials and Methods : Twenty-two patients with head and neck cancer underwent a FDG-PET study before RT, after a flrst dose of 45 Gy, and after a second dose on more 4han 70 Gy. Standard uptake value (SUV) was calculated for primary tumor (n=22) and neck lymph node (n:10). Attenuation corrected PET scans acquired 60 min after tracer injection were used for evaluation of FDG uptake In tumors. A quantitative FDG uptake index was expressed as Suvlean (corrected for iean body mass). The follow-up time was at least 5 months (range S-1 S months). Results : A total of 22 primary tumors and 10 metastatic lymph nodes were analyzed In FDG-PET. In the first PET study the mean SUVlean the primary tumors and nodes were 5.4 (SD, 2.5) and 4.6 (SD, 2.3), respectively. In the second PET, study peformed after 46 Gy RT the mean SUV in primary tumor and node decreased to 2.9 (SD, 1.9, p<0.001) and 1.7 (SD, 1.3) respectively. in the third PET study peformed at the full dose (more than 70 Gy), RT the mean SUV In the primary tumors and nodes decreased to 2.3 (SD, 1.5, p<0.001) and 1.5 (SD, 1 .1) respectively. Conclusions: FDG uptake In tumors showed a significant decrease after the 45 Gy and more than 70 Gy of RT for squamous cell carcinoma of the head and neck. Reduction of metabolic activity after 46 Gy of radiotherapy Is closely correlated with radiation response.

• Fisheries and Aquatic Sciences
• /
• v.26 no.12
• /
• pp.738-751
• /
• 2023

The Sembilan Archipelago is famous for its great biodiversity, in which the humphead wrasse (Cheilinus undulatus) (locally named Napoleon fish) is the primary commodity (economically important), and currently, the environmental degradation occurs due to anthropogenic activities. This study aimed to examine the eco-environmental parameters and assess their influence on the abundance of humphead wrasse and other coral reef fish compositions in the Sembilan Archipelago. Direct field monitoring was performed using a visual census throughout an approximately one km transect. Coral cover data collection and assessment were also carried out. A coastal water quality index (CWQI) was used to assess the water quality status. Furthermore, statistical-based analyses [hierarchical clustering, Pearson's correlation, principal component analysis (PCA), and canonical correspondence analysis (CCA)] were performed to examine the correlation between eco-environmental parameters. The Napoleon fish was only found at stations 1 and 2, with a density of about 3.8 Ind/ha, aligning with the dominant composition of the family Serranidae (covering more than 15% of the total community) and coinciding with the higher coral mortality and lower reef fish abundance. The coral reef conditions were generally ideal for supporting marine life, with a living coral percentage of about > 50% in all stations. Based on CWQI, the study area is categorized as good and excellent water quality. Of the 60 parameter values examined, the phytoplankton abundance, Napoleon fish, and temperature are highly correlated, with a correlation coefficient value greater than 0.7, and statistically significant (F < 0.05). Although the adaptation of reef fish to water quality parameters varies greatly, the most influential parameters in shaping their composition in the study area are living corals, nitrites, ammonia, larval abundance, and temperature.

• Journal of Bio-Environment Control
• /
• v.33 no.2
• /
• pp.79-87
• /
• 2024

This study confirmed the effects of seed shape, temperature, and light treatment on the germination of eight species of medicinal crops to produce high-value crops in vertical farms. Eight species of medicinal seeds were selected, and the seed shape, seed length, seed width, seed length/width ratio, and one hundred seed weight were measured. The seed moisture content was confirmed. Eight species of medicinal seeds were sown, and the germination rate, germination energy, mean daily germination, and mean germination time were investigated according to temperature (15, 20, 25, 25/15℃) and light treatment. Each of the eight medicinal seeds showed different seed shapes. The moisture content of the seeds showed a moisture content rate of over 20% in the five medicinal seeds. Medicinal seeds that showed a germination rate of over 50% were Angelica gigas Nakai, Codonopsis lanceolata (Siebold & Zucc.) Benth. & Hook.f. ex Trautv., and Achyranthes bidentata Blume var. japonica Miq. seeds. A. gigas seed showed a germination rate of 67.34 ± 4.38% under 25/15℃ light conditions, and C. lanceolata seed showed a germination rate of over 50% under both temperature and light treatment conditions, especially the highest germination rate of 82.67 ± 1.46% under 15℃ dark conditions. Peucedanum japonicum Thunb. seed showed a germination rate of 52.34 ± 1.77% under dark conditions at 20℃, and the highest germination rate was 51.67 ± 3.79% under dark conditions at 15℃. The maximum germination energy was 74.00 ± 4.94% in C. lanceolata seed. The maximum mean daily germination was 14.94 ± 0.15 days in P. japonicum seed. Astragalus penduliflorus Lam. var. dahuricus (DC.) X.Y.Zhu seed showed the highest mean germination time of 34.19 ± 4.71. Through this study, it was determined that A. gigas, C. lanceolata, and A. penduliflours seeds would be suitable for production in vertical farms based on the characteristics of each medicinal seed through analysis of seed germination characteristics.

• Journal of Intelligence and Information Systems
• /
• v.23 no.1
• /
• pp.95-108
• /
• 2017

Recently, AlphaGo which is Bakuk (Go) artificial intelligence program by Google DeepMind, had a huge victory against Lee Sedol. Many people thought that machines would not be able to win a man in Go games because the number of paths to make a one move is more than the number of atoms in the universe unlike chess, but the result was the opposite to what people predicted. After the match, artificial intelligence technology was focused as a core technology of the fourth industrial revolution and attracted attentions from various application domains. Especially, deep learning technique have been attracted as a core artificial intelligence technology used in the AlphaGo algorithm. The deep learning technique is already being applied to many problems. Especially, it shows good performance in image recognition field. In addition, it shows good performance in high dimensional data area such as voice, image and natural language, which was difficult to get good performance using existing machine learning techniques. However, in contrast, it is difficult to find deep leaning researches on traditional business data and structured data analysis. In this study, we tried to find out whether the deep learning techniques have been studied so far can be used not only for the recognition of high dimensional data but also for the binary classification problem of traditional business data analysis such as customer churn analysis, marketing response prediction, and default prediction. And we compare the performance of the deep learning techniques with that of traditional artificial neural network models. The experimental data in the paper is the telemarketing response data of a bank in Portugal. It has input variables such as age, occupation, loan status, and the number of previous telemarketing and has a binary target variable that records whether the customer intends to open an account or not. In this study, to evaluate the possibility of utilization of deep learning algorithms and techniques in binary classification problem, we compared the performance of various models using CNN, LSTM algorithm and dropout, which are widely used algorithms and techniques in deep learning, with that of MLP models which is a traditional artificial neural network model. However, since all the network design alternatives can not be tested due to the nature of the artificial neural network, the experiment was conducted based on restricted settings on the number of hidden layers, the number of neurons in the hidden layer, the number of output data (filters), and the application conditions of the dropout technique. The F1 Score was used to evaluate the performance of models to show how well the models work to classify the interesting class instead of the overall accuracy. The detail methods for applying each deep learning technique in the experiment is as follows. The CNN algorithm is a method that reads adjacent values from a specific value and recognizes the features, but it does not matter how close the distance of each business data field is because each field is usually independent. In this experiment, we set the filter size of the CNN algorithm as the number of fields to learn the whole characteristics of the data at once, and added a hidden layer to make decision based on the additional features. For the model having two LSTM layers, the input direction of the second layer is put in reversed position with first layer in order to reduce the influence from the position of each field. In the case of the dropout technique, we set the neurons to disappear with a probability of 0.5 for each hidden layer. The experimental results show that the predicted model with the highest F1 score was the CNN model using the dropout technique, and the next best model was the MLP model with two hidden layers using the dropout technique. In this study, we were able to get some findings as the experiment had proceeded. First, models using dropout techniques have a slightly more conservative prediction than those without dropout techniques, and it generally shows better performance in classification. Second, CNN models show better classification performance than MLP models. This is interesting because it has shown good performance in binary classification problems which it rarely have been applied to, as well as in the fields where it's effectiveness has been proven. Third, the LSTM algorithm seems to be unsuitable for binary classification problems because the training time is too long compared to the performance improvement. From these results, we can confirm that some of the deep learning algorithms can be applied to solve business binary classification problems.

• Journal of Korean Academy of Nursing
• /
• v.25 no.4
• /
• pp.653-667
• /
• 1995

Post - operative wound infections have been a serious problem in nursing care in the operating room and appear to be strongly related to the infection occurring during the performance of operation. The purpose of this study is to identify patterns in duration of surgical hand scrub (SHS), to evaluate the method of SHS and to examine the rate of glove perforation. Subjects for this study include 244 doctors and 169 nurses working in the operative theatre of a hospital in Seoul area. Test samples and related data were collected from this medical facility between April 1, through 15, and July 1, through 5, 1995 by the author and a staff member working in the operating room. For the study, data on the SHS of doctors and nurses were obtained at the time of operation and multiple batches of surgical gloves worn by the operating doctors were collected after each operation. The duration of SHS was measured with a stop watch and the method of SHS was evaluated according to Scoring Hand Scrub Criteria (SHS Criteria) and expressed as SHS scores. For the analysis of the data, t-test was used to compare the differences in the duration and the SHS scores of doctors and nurses, and Pearson's correlation coefficient was used to examine the relationship between the SHS duration and the SHS scores. The results of the study are summarized as follows. 1) The mean time spent in each SHS was 167 seconds in nurses, and 127 seconds in doctors. The data comparing nurses and doctors indicated that there were significant differences in Our ation of SH S between these two groups (t=5.58, p=.000). 2) The mean time spent in the first SHS was 145 seconds and that in the End SHS, 135 seconds, and there was not a significant difference in the duration of the SHS between doctors and nurses (t=1.44, P=.156). 3) The mean time spent in the SHS by OS (Orthopaedic surgery) doctors was 162 seconds, 150 seconds by NS(Neurologic surgery), 121 seconds by GS(General surgery), 94 seconds by OPH(Opthalmology) and DS(Dental surgery), 82 seconds by URO(Urology), 78 seconds by PS(Plastic surgery) and 40 seconds by ENT(Ear, Nose & Throat) These also showed a significant difference in the duration of the SHS among the medical specialities (t=4.8, P=.0001). 4) The average SHS score of the nurses was 15.2, while that of doctors was 13.1. The statistical analysis showed that t-value was 3.66, p was. 000. This indicates that the nurses actually clean their hands more thoroughly than the doctors do. 5) The average SHS score of NS doctors was 15.5, 15.3 for doctors for OPH,14.3 for OS,12.7 for GS, 12.0 for DS, 11.7 for URO, 10.1 for PS, 7.5 for ENT. Comparison of the average SHS scores from 8 specialties showed that there was a significant differences in the patterns of the SHS (F=5.08, P=.000) among medical specialties. 6) It appears that the operating personnel scrub the palms and dorsum of their hand relatively well, however, less thorough the nails and fingers. 7) The more the operating personnel spend their time in hand scrubbing, the more correctly they clean their hands(r=.6427, P<.001). 8) The overall frequencies of perforation in all post-operative gloves tested was 38 out of 389 gloves (10.3%). The perforation rate for PS was 13%, 12.1% for GS,8.8% for 05, and 3.3% for NS.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.40 no.4
• /
• pp.114-126
• /
• 2012

The purpose of this study is to find ways of the view evaluation of biotope classification before development by selecting an area, which is as large as about $10.0km^2$ around Non Gong Up, Auk Po Myun, Dalsung Gun, Daugu where the large project has been planned, as a subject of this study. The results of this study are as follows. Because of the classification of biotope, there are 23 kinds of types that are subdivided into 140 types. Three surveys for selecting the assessment indicators were performed. The first survey analyzed the importance of 22 selected assessment indicators based on the evaluation of an existing literature review and on the spot research. The second survey performed factor analysis and reclassified the value indicators. The third survey computed additive values of the selected assessment indicators. It used a method of standardizing the average importance of indicators by making their sum equal by 10. Theses additive values were then multiplied by each grade of indicators in order to make a final evaluation. The number of assessment indicators finally selected through the survey of asking specialist is vitality elements, visual obstructs elements etc 19. According to the result of evaluation of 1st, 1 grade spaces which especially valuable is analyzed that 7 spaces, 2 grade spaces for 4, 3 grade spaces for 5, 4 grade space for 2, 5 grade space for 5. Because of the evaluation of 2st, 1 grade spaces which especially valuable(1a, 1b) is analyzed that 15 spaces, 2 grade spaces which valuable is analyzed that 28 space. As the evaluation of site suitability model of this study couldn't have high applicability to other similar area because of having only one site as a subject, it is needed to do synthesize and standardization of various examples to have higher objectivity later.