• Journal of Intelligence and Information Systems
• /
• v.21 no.4
• /
• pp.17-35
• /
• 2015

Demand forecasting is the activity of estimating the quantity of a product or service that consumers will purchase for a certain period of time. Developing precise forecasting models are considered important since corporates can make strategic decisions on new markets based on future demand estimated by the models. Many studies have developed market growth curve models, such as Bass, Logistic, Gompertz models, which estimate future demand when a market is in its early stage. Among the models, Bass model, which explains the demand from two types of adopters, innovators and imitators, has been widely used in forecasting. Such models require sufficient demand observations to ensure qualified results. In the beginning of a new market, however, observations are not sufficient for the models to precisely estimate the market's future demand. For this reason, as an alternative, demands guessed from those of most adjacent markets are often used as references in such cases. Reference markets can be those whose products are developed with the same categorical technologies. A market's demand may be expected to have the similar pattern with that of a reference market in case the adoption pattern of a product in the market is determined mainly by the technology related to the product. However, such processes may not always ensure pleasing results because the similarity between markets depends on intuition and/or experience. There are two major drawbacks that human experts cannot effectively handle in this approach. One is the abundance of candidate reference markets to consider, and the other is the difficulty in calculating the similarity between markets. First, there can be too many markets to consider in selecting reference markets. Mostly, markets in the same category in an industrial hierarchy can be reference markets because they are usually based on the similar technologies. However, markets can be classified into different categories even if they are based on the same generic technologies. Therefore, markets in other categories also need to be considered as potential candidates. Next, even domain experts cannot consistently calculate the similarity between markets with their own qualitative standards. The inconsistency implies missing adjacent reference markets, which may lead to the imprecise estimation of future demand. Even though there are no missing reference markets, the new market's parameters can be hardly estimated from the reference markets without quantitative standards. For this reason, this study proposes a case-based expert system that helps experts overcome the drawbacks in discovering referential markets. First, this study proposes the use of Euclidean distance measure to calculate the similarity between markets. Based on their similarities, markets are grouped into clusters. Then, missing markets with the characteristics of the cluster are searched for. Potential candidate reference markets are extracted and recommended to users. After the iteration of these steps, definite reference markets are determined according to the user's selection among those candidates. Then, finally, the new market's parameters are estimated from the reference markets. For this procedure, two techniques are used in the model. One is clustering data mining technique, and the other content-based filtering of recommender systems. The proposed system implemented with those techniques can determine the most adjacent markets based on whether a user accepts candidate markets. Experiments were conducted to validate the usefulness of the system with five ICT experts involved. In the experiments, the experts were given the list of 16 ICT markets whose parameters to be estimated. For each of the markets, the experts estimated its parameters of growth curve models with intuition at first, and then with the system. The comparison of the experiments results show that the estimated parameters are closer when they use the system in comparison with the results when they guessed them without the system.

• Korean Journal of Construction Engineering and Management
• /
• v.15 no.5
• /
• pp.31-39
• /
• 2014

It has been a long time since cars had become important means of transportation in human life. Since 1970s, cars have been increasing steadily because of rising individual income and changing lifestyle toward leisure and convenience. The number of cars is just 1.8 per thousand populations in 1970s, however, in 2012, it has increased to 291.15. Forecasting the demand for cars would be useful to plan, construction or management in the field of motor industry, road building and establishing facilities. Our study predicts the demand of cars through estimating the growth curve model. Especially, we include ageing variables to forecasting identifying the effect of ageing on the demand of cars. The main findings are as follows. In 2045, the number of cars is expected to reach 486.8 per thousand populations with passing a primary saturation point at early 2020s. Also, due to effect of ageing, the predicted demand of cars is about 10% lower than in case of which if ageing effect not exist.

• Investigative Magnetic Resonance Imaging
• /
• v.24 no.1
• /
• pp.38-45
• /
• 2020

Background: Differentiation of cerebellopontine angle (CPA) schwannoma from meningioma is often a difficult process to identify. Purpose: To identify imaging features for distinguishing CPA schwannoma from meningioma and to investigate the usefulness of susceptibility-weighted imaging (SWI) in differentiating them. Materials and Methods: Between March 2010 and January 2015, this study pathologically confirmed 11 meningiomas and 20 schwannomas involving CPA with preoperative SWI were retrospectively reviewed. Generally, the following MRI features were evaluated: 1) maximal diameter on axial image, 2) angle between tumor border and adjacent petrous bone, 3) presence of intratumoral dark signal intensity on SWI, 4) tumor consistency, 5) blood-fluid level, 6) involvement of internal auditory canal (IAC), 7) dural tail, and 8) involvement of adjacent intracranial space. On CT, 1) presence of dilatation of IAC, 2) intratumoral calcification, and 3) adjacent hyperostosis were evaluated. All features were compared using Chi-squared tests and Fisher's exact tests. The univariate and multivariate logistic regression analysis were performed to identify imaging features that differentiate both tumors. Results: The results noted that schwannomas more frequently demonstrated dark spots on SWI (P = 0.025), cystic consistency (P = 0.034), and globular angle (P = 0.008); schwannomas showed more dilatation of internal auditory meatus and lack of calcification (P = 0.008 and P = 0.02, respectively). However, it was shown that dural tail was more common in meningiomas (P < 0.007). In general, dark spots on SWI and dural tail remained significant in multivariate analysis (P = 0.037 and P = 0.012, respectively). In this case, the combination of two features showed a sensitivity and specificity of 80% and 100% respectively, with an area under the receiver operating characteristic curve of 0.9. Conclusion: In conclusion, dark spots on SWI were found to be helpful in differentiating CPA schwannoma from meningioma. It is noted that combining dural tail with dark spots on SWI yielded strong diagnostic value in differentiating both tumors.

• Journal of the Korean Geographical Society
• /
• v.38 no.3
• /
• pp.363-374
• /
• 2003

There is very high preference for the firms to locate in the Capital region, the City of Seoul and its surrounding areas, which inevitably meets diverse types of regulations to prevent over-concentration in Korea. In order to suggest an urgent need to reform these regulations, the demand for knowledge-based industries is estimated. A legit model is employed to estimate the demand of relocation of the current firms based on a survey conducted in 2001. A logistic curve is used to forecast the demand of new start-ups in Korea. The lands for industrial use only are estimated as many as 2.1 million~3.9 million pyung(1 pyung=3.3$m^2$) in nation-wide. Considering affiliate facilities and infrastructures, 3.1 million~5.9 million of industrial area should be developed in Korea for next five years. Since the rents are very high and the available land is short in the southern parts of Seoul, where most knowledge-based firms locate right now. Many firms have considered relocating on any other places where there exist a plenty of lands available and cheaper rents and cheaper wage rates, but still not far away from Seoul so that they could obtain new advanced information, skilled labors, venture capitals, and high quality of producer services. The Capital region, especially Gyeonggi and Incheon, is the only place to meet those conditions in Korea.

• Journal of Korean Society of Rural Planning
• /
• v.21 no.2
• /
• pp.33-49
• /
• 2015

The objective of this study was to predict land use change based on the land use change scenarios for the Hwangguji river watershed, South Korea. The land use change scenario was derived from the representative concentration pathways (RCP) 4.5 and 8.5 scenarios. The CLUE (conversion of land use and its effects) model was used to simulate the land use change. The CLUE is the modeling framework to simulate land use change considering empirically quantified relations between land use types and socioeconomic and biophysical driving factors through dynamical modeling. The Hwangguji river watershed, South Korea was selected as study area. Future land use changes in 2040, 2070, and 2100 were analyzed relative to baseline (2010) under the RCP4.5 and 8.5 scenarios. Binary logistic regressions were carried out to identify the relation between land uses and its driving factors. CN (Curve number) and impervious area based on the RCP4.5 and 8.5 scenarios were calculated and analyzed using the results of future land use changes. The land use change simulation of the RCP4.5 scenario resulted that the area of urban was forecast to increase by 12% and the area of forest was estimated to decrease by 16% between 2010 and 2100. The land use change simulation of the RCP8.5 scenario resulted that the area of urban was forecast to increase by 16% and the area of forest was estimated to decrease by 18% between 2010 and 2100. The values of Kappa and multiple resolution procedure were calculated as 0.61 and 74.03%. CN (III) and impervious area were increased by 0-1 and 0-8% from 2010 to 2100, respectively. The study findings may provide a useful tool for estimating the future land use change, which is an important factor for the future extreme flood.

• Asian Pacific Journal of Cancer Prevention
• /
• v.15 no.18
• /
• pp.7575-7581
• /
• 2014

MicroRNAs (miRNAs) play an essential role in the development and progression of nasopharyngeal carcinomas (NPC). Despite advances in the field of cancer molecular biology and biomarker discovery, the development of clinically validated biomarkers for primary NPC has remained elusive. In this study, we investigated the expression and clinical significance of miRNAs as novel primary NPC diagnostic biomarkers. We used an array containing 2, 500 miRNAs to identify 22 significant miRNAs, and these candidate miRNAs were validated using 67 fresh NPC and 25 normal control tissues via quantitative real-time PCR (qRT-PCR). Expression and correlation analyses were performed with various statistical approaches, in addition to logistic regression and receiver operating characteristic curve analyses to evaluate diagnostic efficacy. qRT-PCR revealed five differentially expressed miRNAs (miR-93-5p, miR-135b-5p, miR-205-5p and miR-183-5p) in NPC tissue samples relative to control samples (p<0.05), with miR-135b-5p and miR-205-5p being of significant diagnostic value (p<0.01). Moreover, comparison of NPC patient clinicopathologic data revealed a negative correlation between miR-93-5p and miR-183-5p expression levels and lymph node status (p<0.05). These findings display an altered expression of many miRNAs in NPC tissues, thus providing information pertinent to pathophysiological and diagnostic research. Ultimately, miR-135b-5p and miR-205-5p may be implicated as novel NPC candidate biomarkers, while miR-93-5p, miR-650 and miR-183-5p may find application as relevant clinical pathology and diagnostic candidate biomarkers.

• Journal of The Korean Society of Clinical Toxicology
• /
• v.14 no.1
• /
• pp.16-25
• /
• 2016

Purpose: Patients suffering from acute poisoning by different substances often visit the emergency department (ED) and receive various prognoses according to the toxic material and patients' condition. Hyperlactatemia, which is an increased blood lactate level that generally indicates tissue hypoperfusion, is commonly utilized as a prognostic marker in critically ill patients such as those with sepsis. This study was conducted to investigate the relationships between blood lactate and clinical prognosis in acute poisoned patients. Methods: This retrospective study was conducted from January 2013 to June 2014 at a single and regional-tertiary ED. We enrolled study patients who were examined for blood test with lactate among acute intoxicated patients. The toxic materials, patient demographics, laboratory data, and mortalities were also reviewed. Additionally, we analyzed variables including blood lactate to verify the correlation with patient mortality. Results: A total of 531 patients were enrolled, including 24 (4.5%) non-survivors. Patient age, Glasgow coma scale (GCS), serum creatinine (Cr), aspartate transaminase (AST), and serum lactate differed significantly between survivors and non-survivors in the binary logistic regression analysis. Among these variables, GCS, AST, and lactate differed significantly. The median serum lactate levels were 2.0 mmol/L among survivors and 6.9 mmol/L among non-survivors. The AUC with the ROC curve and odds ratio of the initial serum lactate were 0.881 and 3.06 (0.89-8.64), respectively. Conclusion: Serum lactate was correlated with fatalities of acute poisoning patients in the ED; therefore, it may be used as a clinical predictor to anticipate their prognoses.

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.28 no.4
• /
• pp.371-379
• /
• 1992

In order to analyse the mesh selectivity for the trawl net, the fishing experiment was carried out by the training ship Saebada in the southern Korea Sea and the East China Sea from June 1991 to August 1992. The trawl net used in experiment has the trouser type of cod-end with cover net, and the mesh selectivity was examined for the five kinds of the opening mesh size in its cod-end part. The selection curves and the selection parameters were calculated by using a logistic function, S=1/(1+exp super(-(aL+b))), and in this case, a and b are the selection parameters and L is the body length of the target species of fishes. In this report, the four species of aquatic animals were analysed because the catch data were enough to calculate normally the selection curves and the selection parameters, and the results obtained are summarized as follows: 1. Trachurus japonicus; Selection parameters a and b in each cases of the opening mesh size of 51.2mm, 70.2mm, 77.6mm, 88.0mm and 111.3mm were respectively 0.5050 and -5.4283, 0.3018 and -4.9590, 0.3816 and -7.3659, 0.2695 and -5.7958, 0.2170 and -5.1226. 2. Photololigo edulis ; Selection Parameters a and b in each cases of the former mesh sizes were respectively 0.7394 and -6.1433, 0.3389 and -4.2366, 0.3286 and -5.1002, 0.2543 and -5.0049, 0.1795 and -4.8040. 3. Trichirus lepturus; Selection curves in the opening mesh size of 111.3mm was calculated unnormally. The selection parameters in the other opening mesh sizes were respectively 0.3790 and -5.2891, 0.2071 and -4.9164, 0.1292 and -3.1733, 0.1153 and -3.8497 in the order of former mesh sizes except 111.3mm. 4. Todarodes pacificus ; Selection curve in case of the opening mesh sizes, 70.2mm and 111.3mm were calculated unnormally. In the order cases of the opening mesh sizes, the selection parameters were respectively were 0.5766 and -6.0169, 0.3735 and -5.4633, 0.2771 and -5.7718 in the order of former mesh sizes except 70.2mm and 111.3mm.

• Tuberculosis and Respiratory Diseases
• /
• v.81 no.3
• /
• pp.247-255
• /
• 2018

Background: Patients with acute respiratory failure secondary to tuberculous destroyed lung (TDL) have a poor prognosis. The aim of the present retrospective study was to develop a mortality prediction model for TDL patients who require mechanical ventilation. Methods: Data from consecutive TDL patients who had received mechanical ventilation at a single university-affiliated tertiary care hospital in Korea were reviewed. Binary logistic regression was used to identify factors predicting intensive care unit (ICU) mortality. A TDL on mechanical Ventilation (TDL-Vent) score was calculated by assigning points to variables according to ${\beta}$ coefficient values. Results: Data from 125 patients were reviewed. A total of 36 patients (29%) died during ICU admission. On the basis of multivariate analysis, the following factors were included in the TDL-Vent score: age ${\geq}65$ years, vasopressor use, and arterial partial pressure of oxygen/fraction of inspired oxygen ratio <180. In a second regression model, a modified score was then calculated by adding brain natriuretic peptide. For TDL-Vent scores 0 to 3, the 60-day mortality rates were 11%, 27%, 30%, and 77%, respectively (p<0.001). For modified TDL-Vent scores 0 to ${\geq}3$, the 60-day mortality rates were 0%, 21%, 33%, and 57%, respectively (p=0.001). For both the TDL-Vent score and the modified TDL-Vent score, the areas under the receiver operating characteristic curve were larger than that of other illness severity scores. Conclusion: The TDL-Vent model identifies TDL patients on mechanical ventilation with a high risk of mortality. Prospective validation studies in larger cohorts are now warranted.

• Asian Pacific Journal of Cancer Prevention
• /
• v.16 no.12
• /
• pp.5095-5099
• /
• 2015

Background: The statistical methods to analyze and predict the related dangerous factors of deep fungal infection in lung cancer patients were several, such as logic regression analysis, meta-analysis, multivariate Cox proportional hazards model analysis, retrospective analysis, and so on, but the results are inconsistent. Materials and Methods: A total of 696 patients with lung cancer were enrolled. The factors were compared employing Student's t-test or the Mann-Whitney test or the Chi-square test and variables that were significantly related to the presence of deep fungal infection selected as candidates for input into the final artificial neural network analysis (ANN) model. The receiver operating characteristic (ROC) and area under curve (AUC) were used to evaluate the performance of the artificial neural network (ANN) model and logistic regression (LR) model. Results: The prevalence of deep fungal infection from lung cancer in this entire study population was 32.04%(223/696), deep fungal infections occur in sputum specimens 44.05%(200/454). The ratio of candida albicans was 86.99% (194/223) in the total fungi. It was demonstrated that older (${\geq}65$ years), use of antibiotics, low serum albumin concentrations (${\leq}37.18g/L$), radiotherapy, surgery, low hemoglobin hyperlipidemia (${\leq}93.67g/L$), long time of hospitalization (${\geq}14$days) were apt to deep fungal infection and the ANN model consisted of the seven factors. The AUC of ANN model($0.829{\pm}0.019$)was higher than that of LR model ($0.756{\pm}0.021$). Conclusions: The artificial neural network model with variables consisting of age, use of antibiotics, serum albumin concentrations, received radiotherapy, received surgery, hemoglobin, time of hospitalization should be useful for predicting the deep fungal infection in lung cancer.