• Journal of Intelligence and Information Systems
• /
• v.25 no.1
• /
• pp.85-107
• /
• 2019

Online consumers browse products belonging to a particular product line or brand for purchase, or simply leave a wide range of navigation without making purchase. The research on the behavior and purchase of online consumers has been steadily progressed, and related services and applications based on behavior data of consumers have been developed in practice. In recent years, customization strategies and recommendation systems of consumers have been utilized due to the development of big data technology, and attempts are being made to optimize users' shopping experience. However, even in such an attempt, it is very unlikely that online consumers will actually be able to visit the website and switch to the purchase stage. This is because online consumers do not just visit the website to purchase products but use and browse the websites differently according to their shopping motives and purposes. Therefore, it is important to analyze various types of visits as well as visits to purchase, which is important for understanding the behaviors of online consumers. In this study, we explored the clustering analysis of session based on click stream data of e-commerce company in order to explain diversity and complexity of search behavior of online consumers and typified search behavior. For the analysis, we converted data points of more than 8 million pages units into visit units' sessions, resulting in a total of over 500,000 website visit sessions. For each visit session, 12 characteristics such as page view, duration, search diversity, and page type concentration were extracted for clustering analysis. Considering the size of the data set, we performed the analysis using the Mini-Batch K-means algorithm, which has advantages in terms of learning speed and efficiency while maintaining the clustering performance similar to that of the clustering algorithm K-means. The most optimized number of clusters was derived from four, and the differences in session unit characteristics and purchasing rates were identified for each cluster. The online consumer visits the website several times and learns about the product and decides the purchase. In order to analyze the purchasing process over several visits of the online consumer, we constructed the visiting sequence data of the consumer based on the navigation patterns in the web site derived clustering analysis. The visit sequence data includes a series of visiting sequences until one purchase is made, and the items constituting one sequence become cluster labels derived from the foregoing. We have separately established a sequence data for consumers who have made purchases and data on visits for consumers who have only explored products without making purchases during the same period of time. And then sequential pattern mining was applied to extract frequent patterns from each sequence data. The minimum support is set to 10%, and frequent patterns consist of a sequence of cluster labels. While there are common derived patterns in both sequence data, there are also frequent patterns derived only from one side of sequence data. We found that the consumers who made purchases through the comparative analysis of the extracted frequent patterns showed the visiting pattern to decide to purchase the product repeatedly while searching for the specific product. The implication of this study is that we analyze the search type of online consumers by using large - scale click stream data and analyze the patterns of them to explain the behavior of purchasing process with data-driven point. Most studies that typology of online consumers have focused on the characteristics of the type and what factors are key in distinguishing that type. In this study, we carried out an analysis to type the behavior of online consumers, and further analyzed what order the types could be organized into one another and become a series of search patterns. In addition, online retailers will be able to try to improve their purchasing conversion through marketing strategies and recommendations for various types of visit and will be able to evaluate the effect of the strategy through changes in consumers' visit patterns.

• Journal of Intelligence and Information Systems
• /
• v.27 no.1
• /
• pp.83-102
• /
• 2021

The government recently announced various policies for developing big-data and artificial intelligence fields to provide a great opportunity to the public with respect to disclosure of high-quality data within public institutions. KSURE(Korea Trade Insurance Corporation) is a major public institution for financial policy in Korea, and thus the company is strongly committed to backing export companies with various systems. Nevertheless, there are still fewer cases of realized business model based on big-data analyses. In this situation, this paper aims to develop a new business model which can be applied to an ex-ante prediction for the likelihood of the insurance accident of credit guarantee. We utilize internal data from KSURE which supports export companies in Korea and apply machine learning models. Then, we conduct performance comparison among the predictive models including Logistic Regression, Random Forest, XGBoost, LightGBM, and DNN(Deep Neural Network). For decades, many researchers have tried to find better models which can help to predict bankruptcy since the ex-ante prediction is crucial for corporate managers, investors, creditors, and other stakeholders. The development of the prediction for financial distress or bankruptcy was originated from Smith(1930), Fitzpatrick(1932), or Merwin(1942). One of the most famous models is the Altman's Z-score model(Altman, 1968) which was based on the multiple discriminant analysis. This model is widely used in both research and practice by this time. The author suggests the score model that utilizes five key financial ratios to predict the probability of bankruptcy in the next two years. Ohlson(1980) introduces logit model to complement some limitations of previous models. Furthermore, Elmer and Borowski(1988) develop and examine a rule-based, automated system which conducts the financial analysis of savings and loans. Since the 1980s, researchers in Korea have started to examine analyses on the prediction of financial distress or bankruptcy. Kim(1987) analyzes financial ratios and develops the prediction model. Also, Han et al.(1995, 1996, 1997, 2003, 2005, 2006) construct the prediction model using various techniques including artificial neural network. Yang(1996) introduces multiple discriminant analysis and logit model. Besides, Kim and Kim(2001) utilize artificial neural network techniques for ex-ante prediction of insolvent enterprises. After that, many scholars have been trying to predict financial distress or bankruptcy more precisely based on diverse models such as Random Forest or SVM. One major distinction of our research from the previous research is that we focus on examining the predicted probability of default for each sample case, not only on investigating the classification accuracy of each model for the entire sample. Most predictive models in this paper show that the level of the accuracy of classification is about 70% based on the entire sample. To be specific, LightGBM model shows the highest accuracy of 71.1% and Logit model indicates the lowest accuracy of 69%. However, we confirm that there are open to multiple interpretations. In the context of the business, we have to put more emphasis on efforts to minimize type 2 error which causes more harmful operating losses for the guaranty company. Thus, we also compare the classification accuracy by splitting predicted probability of the default into ten equal intervals. When we examine the classification accuracy for each interval, Logit model has the highest accuracy of 100% for 0~10% of the predicted probability of the default, however, Logit model has a relatively lower accuracy of 61.5% for 90~100% of the predicted probability of the default. On the other hand, Random Forest, XGBoost, LightGBM, and DNN indicate more desirable results since they indicate a higher level of accuracy for both 0~10% and 90~100% of the predicted probability of the default but have a lower level of accuracy around 50% of the predicted probability of the default. When it comes to the distribution of samples for each predicted probability of the default, both LightGBM and XGBoost models have a relatively large number of samples for both 0~10% and 90~100% of the predicted probability of the default. Although Random Forest model has an advantage with regard to the perspective of classification accuracy with small number of cases, LightGBM or XGBoost could become a more desirable model since they classify large number of cases into the two extreme intervals of the predicted probability of the default, even allowing for their relatively low classification accuracy. Considering the importance of type 2 error and total prediction accuracy, XGBoost and DNN show superior performance. Next, Random Forest and LightGBM show good results, but logistic regression shows the worst performance. However, each predictive model has a comparative advantage in terms of various evaluation standards. For instance, Random Forest model shows almost 100% accuracy for samples which are expected to have a high level of the probability of default. Collectively, we can construct more comprehensive ensemble models which contain multiple classification machine learning models and conduct majority voting for maximizing its overall performance.

• Korean Journal of Heritage: History & Science
• /
• v.55 no.2
• /
• pp.200-211
• /
• 2022

The celadon stools with an openwork ring design which consist of four items as one collection were excavated from Gaeseong, Gyeonggi-do Province. The celadon stools were designated and managed as treasures due to their high arthistorical value in the form of demonstrating the excellence of celadon manufacturing techniques and the fanciful lifestyles during the Goryeo Dynasty. However, one of the items, which appeared to have been repaired and restored in the past, suffered a decline in aesthetic value due to the aging of the treatment materials and the lack of skill on the part of the conservator, raising the need for re-treatment as a result of structural instability. An examination of the conservation condition prior to conservation treatment found structural vulnerabilities because physical damage had been artificially inflicted throughout the area that was rendered defective at the time of manufacturing. The bonded surfaces for the cracked areas and detached fragments did not fit, and these areas and fragments had deteriorated because the adhesive trickled down onto the celadon surface or secondary contaminants, such as dust, were on the adhesive surface. The study identified the position, scope, and conditions of the bonded areas at the cracks UV rays and microscopy in order to investigate the condition of repair and restoration. By conducting Fourier-transform infrared spectroscopy(FT-IR) and portable x-ray fluorescence spectroscopy on the materials used for the former conservation treatment, the study confirmed the use of cellulose resins and epoxy resins as adhesives. Furthermore, the analysis revealed the addition of gypsum(CaSO₄·2H₂O) and bone meal(Ca₁₀ (PO₄)₆(OH)₂) to the adhesive to increase the bonding strength of some of the bonded areas that sustained force. Based on the results of the investigation, the conservation treatment for the artifact would focus on completely dismantling the existing bonded areas and then consolidating vulnerable areas through bonding and restoration. After removing and dismantling the prior adhesive used, the celadon stool was separated into 6 large fragments including the top and bottom, the curved legs, and some of the ring design. After dismantling, the remaining adhesive and contaminants were chemically and physically removed, and a steam cleaner was used to clean the fractured surfaces to increase the bonding efficacy of the re-bonding. The bonding of the artifact involved applying the adhesive differently depending on the bonding area and size. The cyanoacrylate resin Loctite 401 was used on the bonding area that held the positions of the fragments, while the acrylic resin Paraloid B-72 20%(in xylene) was treated on cross sections for reversibility in the areas that provided structural stability before bonding the fragments using the epoxy resin Epo-tek 301-2. For areas that would sustain force, as in the top and bottom, kaolin was added to Epo-tek 301-2 in order to reinforce the bonding strength. For the missing parts of the ring design where a continuous pattern could be assumed, a frame was made using SN-sheets, and the ring design was then modeled and restored by connecting the damaged cross section with Wood epos. Other restoration areas that occurred during bonding were treated by being filled with Wood epos for aesthetic and structural stabilization. Restored and filled areas were color-matched to avoid the feeling of disharmony from differences of texture in case of exhibitions in the future. The investigation and treatment process involving a variety of scientific technology was systematically documented so as to be utilized as basic data for the conservation and maintenance.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.51 no.4
• /
• pp.101-118
• /
• 2023

During the COVID-19 pandemic, the utilization rate of neighborhood parks and green spaces increased significantly, and the outbreak served as an opportunity to highlight the values and functions of neighborhood parks and green spaces for urban residents. This study aims to empirically analyze how citizens' movement and the use of neighborhood parks and green spaces changed before and after COVID-19 and examine the social and spatial characteristics that affected these changes. As a research method, first, people's mobility around neighborhood parks and green spaces before and after the COVID-19 pandemic were compared using signal data from telecommunication carriers. Through the analysis of changes in residence time and movement volume, the movement characteristics of citizens after COVID-19 and changes in walking-based park visits were examined. Second, the factors affecting the mobility change in neighborhood parks and green spaces were analyzed. The social and spatial characteristics that affect citizens' visits to neighborhood parks and green spaces before and after COVID-19 were examined through correlation and multiple regression analysis. Subsequently, through cluster analysis, the types of living areas for the post-COVID era were classified from the perspective of the supply and management of neighborhood parks and green spaces services, and directions for improving neighborhood parks and green spaces by type were presented. Major research findings are as follows: First, since the outbreak of COVID-19, activities within 500m of the residence have increased. The amount of stay and walking movement increased in both 2020 and 2021, which means that the need to review the quantitative standards and attractions of neighborhood parks and green spaces has increased considering the changed scope of the walking and living area. Second, the overall number of visits to neighborhood parks and green spaces by walking has increased since the outbreak of COVID-19. The number of visits to neighborhood parks and green spaces centered on the house and the workplace increased significantly. The park green policy in the post-COVID era should be promoted by discovering underprivileged areas, focusing on areas where residential, commercial, and business facilities are concentrated, and improving neighborhood parks and green services in quantitative and qualitative terms. Third, it was found that the higher the level of park green service, the higher the amount of walking movement. It is necessary to use indicators that contribute to improving citizens' actual park green services, such as walking accessibility, rather than looking at the criteria for securing green areas. Fourth, as a result of cluster analysis, five types of neighborhood parks and green spaces were derived in response to the post-COVID era. This suggests that it is necessary to consider the socioeconomic status and characteristics of living areas and the level of park green services required in future park green policies. This study has academic and policy significance in that it has laid the basis for establishing neighborhood parks and green spaces policy in response to the post-COVID era by using various analysis methodologies such as carrier signal data analysis, GIS analysis, and statistical analysis.

• Korean Journal of Environment and Ecology
• /
• v.37 no.5
• /
• pp.347-366
• /
• 2023

In the present study, the vegetation was classified using the phytosociological method and canonical-correlation analysis (CCA) was implemented to analyze correlation between community structure and environmental factors in the natural habitats of forest byproducts, especially medicinal plants, such as Cudrania tricuspidata, Sorbus commixta, and Hovenia dulcis, in 2021-2022 to provide primary ecological data to establish environmental conditions for wild vegetable cultivation. A total of 11 plots in five regions, 8 plots in three regions, and 17 plots in 5 regions were selected for the natural habitats of C. tricuspidata in southern Korea, S. commixta in high mountains, and H. dulcis in valleys of central Korea, respectively. The importance value in each community was respectively analyzed as follows, in C. tricuspidata community, the importance value of C. tricuspidata (61.10) was the highest, followed by Celtis sinensis, Pinus thunbergii, Neolitsea aciculata, Styrax japonica, Carpinus coreana, Quercus serrata, and Q. acutissima. In Sorbus commixta community, Q. mongolica (57.21) was the highest, followed by, S. commixta (42.58), Betula ermani, Tilia amurensis, A. pseudosieboldianum, A. tschonoskii var. rubripes, Cornus controversa, Magnolia sieboldii, and Taxus cuspidata. In H. dulcis community, H. dulcis (64.58) was the highest, followed by Zelkova serrata, Cornus controversa, A. mono, Q. serrata, C. cordata, and Juglans mandshurica. As the result of the analysis on DBH of the major species having the high importance value, in C. tricuspidata community, C. tricuspidata, C. sinensis, Neolitsea aciculata, and C. coreana show the density of normal distribution, so the dominant status of these species is likely to continue. In S. commixta community, S. commixta show the density of reverse J-shaped curve, so the dominant status of these species is likely to be stable, and Q. mongolica, B. ermani and T. amurensis, show the density of normal distribution, so the dominant status of these species is likely to continue. In H. dulcis community, C. cordata, and J. mandshurica show the density of reverse J-shaped curve, so the dominant status of these species is likely to be stable, and H. dulcis, Z. serrata, C. controversa and A. mono had a formality distribution, suggesting a continuous domination of these species over the other species for the time being. The results of CCA ordination analysis using 11 environmental factors and 30 communities of three taxa classified by TWINSPAN analysis revealed that the altitude showed the strongest correlation with the vegetation. C. tricuspidata community was distributed on the moderate and gentle northeastern slope at low altitude with the highest pH, C.E.C, Ca²⁺, and Mg² and various P₂O₅, whereas S. commixta community was distributed on the steep slope at high altitude with the highest O.M and T-N and lower P₂O₅, Ca²⁺, Mg²⁺, C.E.C and pH, which is the opposite tendency of the environment of C. tricuspidata community. H. dulcis community was distributed on the gentle northern slope at lower altitude with an average pH, O.M, T-N, Ca²⁺, Mg²⁺, and C.E.C, except higher P₂O₅.

• Bangmulgwan gwa yeongu (The National Museum of Korea Journal)
• /
• v.1
• /
• pp.36-73
• /
• 2024

The Tomb of Jang Mui located in Bongsan-gun, Hwanghae-do Province has attracted much attention since its first investigation due to the inscribed bricks found there that have allowed the guessing of the name and official title of its occupant and construction date. Inscriptions on these bricks, such as the "Prefect of Daebang Commandery Jang Mui" and the "Mu" (戊, the fifth sign of the Chinese calendar), and "Sin" (申, the ninth sign of the Chinese zodiac), have become the basis for believing the location of the government office of Daebang Commandery to be in Bongsangun, Hwanghae-do Province rather than somewhere in the Hangang River region. From the early days of its investigation, the tomb was suggested as historic remains of the Daebang Commandery along with the Earthen Fortress in Jitap-ri. Inscribed bricks excavated from the Tomb of Jang Mui were featured in several books and articles in the form of photographs and rubbings, leading to a vast body of studies on its construction period and the characteristics of its occupant that drew upon interpretations of the inscriptions. However, the inscribed bricks themselves were not publicly available outside those held in the collection of the University of Tokyo, making it difficult to expect consistent research findings on the types of inscribed bricks and their contents. Following previous studies re-examining the structure of the tomb and the materials used for its construction, most scholars dated the Tomb of Jang Mui to 348, a period after the collapse of Daebang Commandery. However, there is still a lack of adequate examination of the bricks, which account for the majority of the artifacts excavated from the tomb. Among the bricks excavated from most brick chamber tombs, including the Tomb of Jang Mui, only those with inscriptions or designs have been collected. Moreover, among these, only those with inscriptions or designs on the stretcher faces have been documented. Accordingly, the bricks themselves have been notably understudied. This paper intends to reorganize the contents of the inscriptions on eleven types (out of sixty-one pieces) of bricks in the collection of the National Museum of Korea, which make up the majority of the bricks excavated from the Tomb of Jang Mui. It also classified them according to their shapes. Furthermore, it examined the bricks from the Tomb of Jang Mui as architectural materials by focusing on their production techniques, including their forming, drying, and firing. Taking a more specific approach, it then compared the results to other bricks from the second century through the fourth century: those from the brick chamber tombs of the Nangnang and Daebang Commanderies and those from the brick chamber tombs built after Nangnang and Daebang Commanderies were ousted. The examination of bricks from the Tomb of Jang Mui has revealed that these bricks were basically produced using the brick manufacturing techniques of Nangnang, but they incorporated new elements found in bricks from brick chamber tombs or brick-and-stone chamber tombs constructed around the mid-fourth century in terms of their size, the use of lime, and the number of inscribed bricks. This supports the prevailing view that the date of the construction of the Tomb of Jang Mui is 348. The Tomb of Jang Mui sustained the existing brick chamber tomb burial tradition, but its ceiling was finished with stone. It demonstrates a blending of the brick chamber tomb practice of the Nangnang and Daebang Commanderies by using bricks produced based on related techniques, but with new elements such as the addition of a lime layer to the bricks. This fusion reflects the political circumstances of its time, such as the expulsion of the Daebang Commandery and the advance of the Goguryeo Kingdom, leading to diverse interpretations. Given archaeological evidence such as the structure, materials, and location of the tomb, the Tomb of Jang Mui appears to be highly related to the Goguryeo Kingdom. However, the forms of the inscribed bricks and the contents of the inscriptions share similarities with brick chamber tombs constructed during the third and fourth centuries in the Jiangsu and Zhejiang regions in China. Further studies on whether the use of lime was an influence from Goguryeo or a continuation of the Daebang tradition and a comparative examination with contemporaneous stone ceiling tombs will provide a more refined understanding of the Tomb of Jang Mui.

• Korean Chemical Engineering Research
• /
• v.45 no.2
• /
• pp.160-165
• /
• 2007

The synthesis and the characterization of Low Silica X (LSX) zeolite for nitrogen adsorption have been studied. The performance of LSX zeolite for nitrogen adsorption was compared to that of the commercial zeolite. The $Na_2O/(Na_2O+K_2O)$ ratio in the gel and the crystallization time were fixed as the synthetic factor. The LSX zeolite was formed at the $Na_2O/(Na_2O+K_2O)$ ratio of 0.75. The formation of LSX zeolite was confirmed by XRD and SEM. The Si/Al ratio was investigated by using XRF and FT-IR. The synthesized LSX zeolite showed a lower Si/Al ratio than the NaY and NaX zeolites although they have a same faujasite structure. The Si/Al ratio of the LSX zeolite converged close to 1. 1A (Li, Na, K) and 2A (Mg, Ca, Ba) group elements were ion-exchanged to the LSX zeolite. As the charge density of cation rises, the amount of nitrogen adsorbed increased. $Li^+$ ion-exchanged LSX zeolite showed the highest nitrogen adsorption weight. When the Li/Al ratio was over 0.65, nitrogen adsorption increased remarkably. $Li^+$ ions located on the supercage (site III, III') in the LSX zeolite played a role as nitrogen adsorption sites. When the $Ca^{2+}$ ions were added to the LiLSX zeolite by ion-exchange method, the performance for nitrogen adsorption increased more. The performance for the nitrogen adsorption was the highest at the Ca/Al ratio of 0.26. Nitrogen adsorption capacity of LiCaLSX (Ca/Al=0.26) zeolite was superior to the commercial NaX zeolite.

• Clinical and Experimental Pediatrics
• /
• v.49 no.9
• /
• pp.977-982
• /
• 2006

Purpose : There has been an increasing amount of literature concerning the association between Mycoplasma pneumoniae and asthma pathogenesis. Interleukin(IL)-6 stimulates the differentiation of monocytes, and can promote Th2 differentiation and simultaneously inhibit Th1 polarization. IL-8 is a potent chemoattractant and, it has been suggested, has a role in asthma pathogenesis. Nitric oxide (NO) synthesized by airway epithelium may be important in the regulation of airway inflammation and reactivity. Vascular endothelial growth factor(VEGF) has been reported to be a mediator of airway remodeling in asthma. We investigated the effects of M. pneumoniae on IL-6, IL-8, NO and VEGF production in human respiratory epithelial cells. Methods : A549 cells were cultured and inoculated with M. pneumoniae at a dose of 20 cfu/cell. After infection, the presence of M. pneumoniae in epithelial cell cultures was monitored by immunofluorescence and confirmed by polymerase chain reaction(PCR) detection. IL-6, IL-8 and VEGF were determined by an enzyme-linked immunosorbent assay and reverse transcriptase-polymerase chain reaction. NO was measured using the standard Griess reaction. Results : In A549 cells, M. pneumoniaeinduced IL-6, IL-8, NO and VEGF release in time-dependent manners. It also induced mRNA expression of IL-6, IL-8 and VEGF in similar manners. Conclusion : These observations suggest that M. pneumoniae might have a role in the pathogenesis of the allergic inflammation of bronchial asthma.

• Clinical and Experimental Pediatrics
• /
• v.49 no.9
• /
• pp.946-951
• /
• 2006

Purpose : The purpose of this study was to demonstrate the effectiveness of combined exercise for 12 weeks on the adiponectin and obesity related variables in overweight and obese children. Methods : Eighteen children in 5th grade in a certain elementary school in Busan were recruited. They were all overweight or obese children(more than 85 percentile in body mass index). Nine children in the experimental group were given exercises consisting of walking and band resistant training for 12 weeks. Auxological data(including height, weight and body fat mass) and laboratory data (fasting blood sugar, insulin, adiponectin) were checked at baseline and at the 1 week, and at the 4 weeks and 12 weeks stages of their exercise program. Insulin resistance and sensitivity were evaluated indirectly using HOMA index and QUICKI index. Results : Adiponectin gradually decreased until the 4 weeks point and gradually increased thereafter to the starting level at the 12 weeks stage. Body weight, body mass index(BMI) and HOMA index significantly decreased more at the 1 week, 4 weeks, and 12 weeks stages in the experimental group than in the control group. Body fat mass significantly decreased at 12 weeks. The change of insulin was significantly correlated with changes of body weight and BMI. But there was no correlation between changes of adiponectin and changes of insulin. Conclusion : Exercise seems to effect the adiponectin concentration. And it might be assumed that exercise increases the adiponectin concentration if it is continued for long time(may be more than 12 weeks). More studies may be necessary to draw that conclusion.

• Journal of Soil and Groundwater Environment
• /
• v.13 no.1
• /
• pp.77-91
• /
• 2008

In this study, we developed a technique of applying DRASTIC, which is the most widely used tool for estimation of groundwater vulnerability to the aqueous phase contaminant infiltrated from the surface, and a groundwater flow model jointly to assess groundwater contamination potential. The developed technique is then applied to Buyeo-eup area in Buyeo-gun, Chungcheongnam-do, Korea. The input thematic data of a depth to water required in DRASTIC model is known to be the most sensitive to the output while only a few observations at a few time schedules are generally available. To overcome this practical shortcoming, both steady-state and transient groundwater level distributions are simulated using a finite difference numerical model, MODFLOW. In the application for the assessment of groundwater vulnerability, it is found that the vulnerability results from the numerical simulation of a groundwater level is much more practical compared to cokriging methods. Those advantages are, first, the results from the simulation enable a practitioner to see the temporally comprehensive vulnerabilities. The second merit of the technique is that the method considers wide variety of engaging data such as field-observed hydrogeologic parameters as well as geographic relief. The depth to water generated through geostatistical methods in the conventional method is unable to incorporate temporally variable data, that is, the seasonal variation of a recharge rate. As a result, we found that the vulnerability out of both the geostatistical method and the steady-state groundwater flow simulation are in similar patterns. By applying the transient simulation results to DRASTIC model, we also found that the vulnerability shows sharp seasonal variation due to the change of groundwater recharge. The change of the vulnerability is found to be most peculiar during summer with the highest recharge rate and winter with the lowest. Our research indicates that numerical modeling can be a useful tool for temporal as well as spatial interpolation of the depth to water when the number of the observed data is inadequate for the vulnerability assessments through the conventional techniques.