• Journal of Intelligence and Information Systems
• /
• v.18 no.1
• /
• pp.23-38
• /
• 2012

There is a large difference between purchasing patterns in an online shopping mall and in an offline market. This difference may be caused mainly by the difference in accessibility of online and offline markets. It means that an interval between the initial purchasing decision and its realization appears to be relatively short in an online shopping mall, because a customer can make an order immediately. Because of the short interval between a purchasing decision and its realization, an online shopping mall transaction usually contains fewer items than that of an offline market. In an offline market, customers usually keep some items in mind and buy them all at once a few days after deciding to buy them, instead of buying each item individually and immediately. On the contrary, more than 70% of online shopping mall transactions contain only one item. This statistic implies that traditional data mining techniques cannot be directly applied to online market analysis, because hardly any association rules can survive with an acceptable level of Support because of too many Null Transactions. Most market basket analyses on online shopping mall transactions, therefore, have been performed by expanding the co-occurrence criteria of traditional association rule mining. While the traditional co-occurrence criteria defines items purchased in one transaction as concurrently purchased items, the expanded co-occurrence criteria regards items purchased by a customer during some predefined period (e.g., a day) as concurrently purchased items. In studies using expanded co-occurrence criteria, however, the criteria has been defined arbitrarily by researchers without any theoretical grounds or agreement. The lack of clear grounds of adopting a certain co-occurrence criteria degrades the reliability of the analytical results. Moreover, it is hard to derive new meaningful findings by combining the outcomes of previous individual studies. In this paper, we attempt to compare expanded co-occurrence criteria and propose a guideline for selecting an appropriate one. First of all, we compare the accuracy of association rules discovered according to various co-occurrence criteria. By doing this experiment we expect that we can provide a guideline for selecting appropriate co-occurrence criteria that corresponds to the purpose of the analysis. Additionally, we will perform similar experiments with several groups of customers that are segmented by each customer's average duration between orders. By this experiment, we attempt to discover the relationship between the optimal co-occurrence criteria and the customer's average duration between orders. Finally, by a series of experiments, we expect that we can provide basic guidelines for developing customized recommendation systems. Our experiments use a real dataset acquired from one of the largest internet shopping malls in Korea. We use 66,278 transactions of 3,847 customers conducted during the last two years. Overall results show that the accuracy of association rules of frequent shoppers (whose average duration between orders is relatively short) is higher than that of causal shoppers. In addition we discover that with frequent shoppers, the accuracy of association rules appears very high when the co-occurrence criteria of the training set corresponds to the validation set (i.e., target set). It implies that the co-occurrence criteria of frequent shoppers should be set according to the application purpose period. For example, an analyzer should use a day as a co-occurrence criterion if he/she wants to offer a coupon valid only for a day to potential customers who will use the coupon. On the contrary, an analyzer should use a month as a co-occurrence criterion if he/she wants to publish a coupon book that can be used for a month. In the case of causal shoppers, the accuracy of association rules appears to not be affected by the period of the application purposes. The accuracy of the causal shoppers' association rules becomes higher when the longer co-occurrence criterion has been adopted. It implies that an analyzer has to set the co-occurrence criterion for as long as possible, regardless of the application purpose period.

• Journal of Intelligence and Information Systems
• /
• v.16 no.3
• /
• pp.99-120
• /
• 2010

Due to the wide spread of customers' frequent access of non face-to-face services, there have been many attempts to improve customer satisfaction using huge amounts of data accumulated throughnon face-to-face channels. Usually, a call center is regarded to be one of the most representative non-faced channels. Therefore, it is important that a call center has enough agents to offer high level customer satisfaction. However, managing too many agents would increase the operational costs of a call center by increasing labor costs. Therefore, predicting and calculating the appropriate size of human resources of a call center is one of the most critical success factors of call center management. For this reason, most call centers are currently establishing a department of WFM(Work Force Management) to estimate the appropriate number of agents and to direct much effort to predict the volume of inbound calls. In real world applications, inbound call prediction is usually performed based on the intuition and experience of a domain expert. In other words, a domain expert usually predicts the volume of calls by calculating the average call of some periods and adjusting the average according tohis/her subjective estimation. However, this kind of approach has radical limitations in that the result of prediction might be strongly affected by the expert's personal experience and competence. It is often the case that a domain expert may predict inbound calls quite differently from anotherif the two experts have mutually different opinions on selecting influential variables and priorities among the variables. Moreover, it is almost impossible to logically clarify the process of expert's subjective prediction. Currently, to overcome the limitations of subjective call prediction, most call centers are adopting a WFMS(Workforce Management System) package in which expert's best practices are systemized. With WFMS, a user can predict the volume of calls by calculating the average call of each day of the week, excluding some eventful days. However, WFMS costs too much capital during the early stage of system establishment. Moreover, it is hard to reflect new information ontothe system when some factors affecting the amount of calls have been changed. In this paper, we attempt to devise a new model for predicting inbound calls that is not only based on theoretical background but also easily applicable to real world applications. Our model was mainly developed by the interactive decision tree technique, one of the most popular techniques in data mining. Therefore, we expect that our model can predict inbound calls automatically based on historical data, and it can utilize expert's domain knowledge during the process of tree construction. To analyze the accuracy of our model, we performed intensive experiments on a real case of one of the largest car insurance companies in Korea. In the case study, the prediction accuracy of the devised two models and traditional WFMS are analyzed with respect to the various error rates allowable. The experiments reveal that our data mining-based two models outperform WFMS in terms of predicting the amount of accident calls and fault calls in most experimental situations examined.

• Restorative Dentistry and Endodontics
• /
• v.32 no.4
• /
• pp.343-355
• /
• 2007

The objectives of this study was to evaluate the durability of 4 resin cements by means of microtensile bond strength test combined with thermocycling method and fractographic FE-SEM analysis. Experimental groups were prepared according to thermocycling (0, 1,000, 5,000) and the kind of resin cements, those were Variolink II, Multilink, Panavia F 2.0, Rely X Unicem. Flat dentin surfaces were created on mid-coronal dentin of extracted third molars. Then fresh dentin surface was grounded with 320-grit silicon carbide abrasive papers to create uniform smear layers. Indirect composite block (Tescera, Bisco Inc., Schaumburg, IL, USA) was fabricated ($12\;{\times}\;12\;{\times}\;6\;mm^3$). It's surface for bonding to tooth was grounded with silicon carbide abrasive papers from 180- to 600-grit serially, then sandblasted witk $20\;-\;50\;{\mu}m$ alumina oxide. According to each manufacturer's instruction, dentin surface was treated and indirect composite block was luted on it using each resin cement. For Rely X Unicem, dentin surface was not treated. The bonded tooth-resin block were stored in distilled water at $37^{\circ}C$ for 24 hours. After thermocycling, the bonded tooth-resin block was sectioned occluso-gingivally to 1.0 mm thick serial slabs using all Isomet slow-speed saw (Isomet, Buehler Ltd, Lake Bluff, IL, USA). These sectioned slabs were further sectioned to $1.0\;{\times}\;1.0\;mm^2$ composite-dentin beams. The specimens were tested with universal testing machine (EZ-Test, Shimadzu, Japan) at a crosshead speed of 1.0 mm/min with maximum load of 500 N. The data was analyzed using one-way ANOVA and Duncan's multiple comparison test at $p\;{\leq}\;0.05$ level. Within the limited results, we conclude as follows; 1. The bond strength of Variolink II was evaluated the highest among experimental groups and was significantly decreased after 1,000 thermocycling (p < 0.05). 2. The bond strength of Multilink was more affected by thermocycling than the other experimental groups and significantly decreased after 1,000 thermocycling (p < 0.05). 3. Panavia F 2.0 and Rely X Unicem showed the gradually decreased tendency of microtensile bond strength according to thermocycling but there was no significant difference (p > 0.05). 4. Adhesive based-resin cements showed lower bond strength with or without thermocycling than composite based-resin cements. 5. Variolink II & Multilink showed high bond strength and mixed failure, which was occurred with a thin layer of luting resin cement before thermocycling and gradually increased adhesive failure along the dentin surface after thermocycling. The bonding performance of resin cement can be affected by application procedure and chemical composition. Composite based-resin cement showed higher bond strength and durability than adhesive based-resin cement.

• Journal of the Korean Society of Food Science and Nutrition
• /
• v.37 no.4
• /
• pp.459-464
• /
• 2008

This study was performed to investigate the effect of calcium-rich large anchovy on calcium metabolism in rats for 5 weeks. Experimental animals were randomly assigned to 5 treatments with 14 heads of Spraque Dawley male rats in each group. The experimental diets were as follows; market milk group (M) as control, market milk+calcium-rich large anchovy group (MA), market milk＋calcium carbonate group (MC), market milk+calcium lactate group (ML), and enriched-calcium market milk group (M2), which were formulated with commercially semi-purified rat chow (AIN-diet) to maintain the same level of calcium (1%) in all groups. Femur lengths of M and M2 groups were significantly higher than other groups. Bone mineral density (BMD) and bone mineral content (BMC) and calcium content of femur were the highest in MA group than other groups. In vitro and in vivo calcium absorption rates were high in MA group (7.30% vs 27.50%) compared with those of the other groups. Serum total-cholesterol and HDL-cholesterol levels were significantly different between M group and MA group (p<0.05). Creatinine levels were significantly higher in M, MA and MC groups than in M2 group (p<0.05). Serum calcium, osteocalcin and ALPase activities were higher in calcium-rich large anchovy (MA) group among the treatments, but there was no significant difference. SGOT activity was significantly lower in M2 group than those of M, MA and MC groups (p<0.05). These results may indicate that the calcium-rich large anchovy has enforced the BMD, BMC and calcium absorption rates of in vitro and in vivo compared with the other groups and might be a calcium-enriched food with large anchovy.

• Journal of Korean Society of Steel Construction
• /
• v.23 no.4
• /
• pp.503-514
• /
• 2011

In this research, the seismic connection details for two concrete-filled U-shape steel beam-to-H columns were proposed and cyclically tested under a full-scale cruciform configuration. The key connecting components included the U-shape steel section (450 and 550 mm deep for specimens A and B, respectively), a concrete floor slab with a ribbed deck (165 mm deep for both specimens), welded couplers and rebars for negative moment transfer, and shear studs for full composite action and strengthening plates. Considering the unique constructional nature of the proposed connection, the critical limit states, such as the weld fracture, anchorage failure of the welded coupler, local buckling, concrete crushing, and rebar buckling, were carefully addressed in the specimen design. The test results showed that the connection details and design methods proposed in this study can well control the critical limit states mentioned above. Especially, the proposed connection according to the strengthening strategy successfully pushed the plastic hinge to the tip of the strengthened zone, as intended in the design, and was very effective in protecting the more vulnerable beam-to-column welded joint. The maximum story drift capacities of 6.0 and 6.8% radians were achieved in specimens A and B, respectively, thus far exceeding the minimumlimit of 4% radians required of special moment frames. Low-cycle fatigue fracture across the beam bottom flange at a 6% drift level was the final failure mode of specimen A. Specimen B failed through the fracture of the top splice plate of the bolted splice at a very high drift ratio of 8.0% radian.

• The Journal of Korean Physical Therapy
• /
• v.16 no.1
• /
• pp.140-156
• /
• 2004

This study, to find whether a sports activity and a speciality aptitude activity in school-childhood can affect in forming E.Q, has been done in Gyoung-Nam area and Busan wide city with asked 222 of men and women being in their school-childhood and a group of 85 people who had not joined in such activity, using a E.Q testing paper provided by Dae-Gyo Education Corp,. and Seoul National University Educational Research Institute. Following is the results after analyzing the compiled datas. 1. The E.Q. level difference between people who joined, and who not joined in a sports action activity was reported existing, the total E.Q average of those who joined was resulted 212.6, a point 29.6 higher than those not joined of 183.0 ( p< .05). As for the E.Q causing points, it resulted that the points of the joined group is generally up than that of the non-joined group, especially this difference was remarkable in terms of feeling recognition or feeling control, a finding that deserves an attention ( p<.05) 2. Joining periods of sports activity did also have relation to develping E.Q. of school-childhood according to this research, the total E.Q points of a group joined in the activity more than 2 years was 215.5 points, which was 17.4 points higher than those not joined of 186.5 points ( p< .05). Backing again to E.Q causing points in this case, it resulted without exeption of all main causes that those who joined in more than 2 years are generally higher than that of those joined less than 2 years, especially the difference was regarded as big in terms of feeling recognition or feeling control, a finding that deserves an attention ( p<.01). 3. The E.Q. differnce between those joined in a specialty aptitude activity and not joined was studied existing, the total E.Q average points of those joined in a specialty aptitude activity was 207.8, a higher figure by 21.3 points than those not joined group of 186.5 ( p< .05). As for the E.Q causing points, it resulted without exeption of all main causes that those who joined are generally higher than that of those not joined, especially for feeling recognition or feeling control, this difference was more clear, a finding that deserves an attention ( p<.01). 4. It also resulted that E.Q growth depends on the periods to have joined in a speciality aptitude activity, for example, the total E.Q points of those joined in the activity more than 2 years was 217.1, a total more higher by 13.5 points than 203.6 of those not joined ( p< .05). For the E.Q. causing points, it, with the exception of empathy was resulted that those who joined in the speciality aptitude activity more than 2 years are generally higher than those joined less than 2 years, especially the difference is remarkable in terms of feeling recognition or feeling control, a finding that is also remarkable ( p<.05). 5. The E.Q difference between the men and women who joined in both activities of sports & speciality aptitude was found existing, the total E.Q. average for women was resulted 214.2 points, which was 9.2 points higher than men of 205.0. As for the E.Q. causing points, which, without exeption of main causes, women's was reported being high than that of men, in special is more remarkable in terms of feeling control, a finding that deserves an attention. ( p<.05).

• The Journal of Korean Physical Therapy
• /
• v.3 no.1
• /
• pp.151-174
• /
• 1991

The purpose of this study is to research the actual conditions of industrial accident patients and to produce worker satisfaction and a rational and effective counter measure plan. Direct interviews with 179 cases (in and out patients) were carried out during a three month period from April to July 1990, at six hospitals : two general hospitals Sun Lin and Sung Mo in Po Hang, and four general hospitals in Taegu : Kyung pooh University Hospital, Dong San Medical Center, Young Nam Medical Center and Catholic Hospital. The results of this study are summarized as fellows : 1. Among the 179 cases, $51.6\%$ were male and $48.4\%$ were female. The two largest age groups were 30-39, $31.8\%$ and 20-29, $27.4\%$. Among the 179 cases, $51.6\%$ were married, the largest family number was 2 to 3, $41.1\%$ and 4 to 5, $25.6\%$. Educationally, graduation from high school was the largest group, $46.4\%$ among ,the patients, followed by middle school and primary school. The largest group income level was from 40-69 만원, $45.2\%$. The largest group of patients who worked over 50 hrs. a week was $52.0\%$. The largest group of patients who worked less than 1 year was $44.7\%$, of the patients in work places of less than 100 people, $60.3\%$ were injured and in work places of 100-299 people, $20.1\%$ were injured. In manufacturing, the largest group injured was $55.3\%$, the next group was transport, storage, communication. The largest group of production workers injured was $40.2\%$. 2. The cause of injury in the largest group was facility problems, $33.5\%$. The next group was unsafe habits, $30.2\%$ ; a lack of safety knowledge, $17.9\%$ ; and insufficient supervision, $12.3\%$. The 30-39 year age group head the highest number of injuries, $40.4\%$ ; work places with more than 10 years of work, $44.4\%$ ; work places with more than 1000 people, $56.3\%$ and mining accidents, $80.0\%$. Among. these groups the highest cause of injury was due to facility problems. 3. The accident pattern showed machinery injuries $28.5\%$ as the largest group, followed by falls & falling objects $17.3\%$, fire & electric $15.1\%$, strucke by an object $14.5\%$, followed by overaction and vehicular accidents. The accident pattern showed $46.4\%$ among workers over the 50 year age group, workers in the 5-10 year group, $50.0\%$ ; places employing more than 1000 workers, $35.3\%$ ; construction $73.7\%$, and construction workers $57.1\%$, among these fall & falling objects caused the greatest number of injuries. 4. The largest group of injuries was fractures $54.8\%$, trauma $14.5\%$, amputation $11.7\%$, open wound, and burns. The largest number of fractures occurred in people in the 30-39 year age group, $63.2\%$ : over 10 years of work, $55.0\%$ ; in work places of 300-490 people, $63.6\%$ ; construction $63.2\%$ and general workers $57.2\%$. 5. The largest group of injuries was upper extremity $45.3\%$, lower extremity $24.0\%$, trunk $18.5\%$ and head or neck $12.2\%$. Of these groups, upper extremity injuries were the highest in those less than 20 years old $75.0\%$, less than 1 year or work $59.5\%$, in work places of 500-999 people $60.0\%$, manufacturing $56.6\%$ and production workers $55.6\%$. 6. Periods of injury showed 34 people injured in September, to be the largest followed by October, 32 ; August, 22 people : July, 19 people and the lowest December, 2 people. During the week, Friday had the largest group injured, 35 people ; followed by Saturday, 26 people and the lowest was Wednesday, 17 people, During the day 1400 hours had the largest group injured, 38 people ; followed by 800 hours, 31 people. 7. On a basis of 5 as the highest mark, the average, according to worker satisfaction showed facility safety 3.55, work environment 3.47, income 3.44, job 3.21 and treatment 2.98. 8. The correlation between general characteristics and injury showed that age was directly correlated to the duration of work (r=2591) p<0.01, age was directly correlated to industry (r=2311) p<0.01, and the duration was directly correlated to occupation (r=4372) p<0.001.

• Journal of Chest Surgery
• /
• v.38 no.2 s.247
• /
• pp.110-115
• /
• 2005

Background: The aims of this paper were to review the mid term clinical results and to analyze the preoperative risk factors of isolated aortic valve replacement (AVR). Material and Method: Between January 1992 and February 2003, 80 patients underwent isolated AVR. 58 were male and 22 were female patients, raging from 12 to 75 years of age (mean :$46.8{\pm}13.0$ years). 74 patients except one early death and 5 follow-up loss were contacted by OPD or by telephone. The mean duration of follow-up was $44.2{\pm}29.7$ months and the total cumulative period was 272.8 patient-year. Result: The complications in hospital occurred in 35 cases : 12 wound problems (11 superficial, 1 deep), 11 arrhythmias (9 temporary, 2 persistent), 3 low cardiac output, and so forth. The late deaths were 4 cases : the heart-related deaths were 2 cases ($0.7\%$ patient-year). Conclusion: The risk factors that influenced the early mortality and morbidity were older age (> 60 years)(p=0.04), poor preoperative NYHA functional class (> 3) (p=0.048), high preoperative serum creatinin level (> 1.2 mg/100 ml)(p=0.031), long operation time (aortic clamping time>90 min)(p=0.042). The same factors influenced the late mortality and morbidity. Freedom from valve-related complication was $86.4{\pm}5.3\%,$ actuarial survival rate were $96.8{\pm}2.3\%$ at 3 years and $90.8{\pm}4.6\%$ at 10 years.

• Journal of Chest Surgery
• /
• v.40 no.1 s.270
• /
• pp.32-36
• /
• 2007

Background: Although acute renal failure (ARF) after coronary artery bypass graft (CABG) is relatively rare, but devastating complication with high mortality. Our study aims to evaluate the effectiveness of early application of CRRT in patients with ARF which developed after on-pump CABG. Material and Method: Two hundred and eighty seven patients underwent isolated on-pump CABG between May 2002 and Feb. 2006 at our institution, of whom 15 (5.2%) needed CRRT (11 patients for postoperatively developed ARF and the remaining 4 patients with preexisting dialysis-dependent chronic renal failure (CRF) for postoperative hemodynamic and metabolic control). Criteria for early application of CRRT were as follows; decreased urine output less than 0.5cc/h/kg for 2 consecutive hours and elevated serum creatinine level greater than 2.0 mg/dL. Result: The incidence of ARF requiring CRRT after on-pump CABG was 3.9% (11/283) and the overall hospital mortality of patient with CRRT was 33.3% (5/15). Of 5 deaths, 4 were patients with postoperatively developed ARF, and 1 was a patient with pre-existing dialysis-dependent CRF patient. The mean time between the operation and the initiation of CRRT was $25.8{\pm}5.8$ hours and the mean duration of CRRT was $62.1{\pm}41.2$ hours. Of the 7 survivors who were not on dialysis-dependent preoperatively, 6 patients fully recovered renal function during hospital stay and 1 patient required permanent renal supportive treatment after discharge from hospital. Conclusion: Early application of CRRT could maintain stable postoperative hemodynamic status and make outcomes better than those of previous reports in patients with ARF which developed after on-pump CABG.

• Land and Housing Review
• /
• v.2 no.4
• /
• pp.517-528
• /
• 2011

Indiscreet developments cause environmental problems in major cities of Seoul Metropolitan Area. Among the environmental problems, the air pollution leads the citizens' physical and economic damages. Therefore, it needs to predict how much air pollutant which is emitted from human activities can be carried by urban environment, then to examine the reasonable level of urban development This study assumed that the air pollution is represented differently by the amount of emission. With the assumption, the acceptable air pollutant emission which keeps the air quality under the environmental standard is estimated, then the proper population is calculated in the case of Gwacheon, Gyeonggi. The result is as follow: First, air pollution concentrations of CO, $NO_2$, $SO_2$ which are estimated by using IDW interpolation of GIS don't excess the air environmental standard. Second, the result of correlation analysis between air pollutant emission and air pollution concentration shows that CO and $NO_2$ has high correlationship with total source of pollution and linear source of pollution, and $SO_2$ with linear source of pollution. Third, the results of regression analysis show that the acceptable population is bigger that the real population in the case of CO, and with the estimation of $NO_2$ and $SO_2$, the current population in the urban center and boundaries where the residential and commerce land uses are concentrated is bigger than the acceptable population. The consequence of this study is that the estimation of carrying capacity can suggest the acceptable human activities which keep the air quality under the environmental standard. This can leads the sustainable urban development by control the human activities under the carrying capacity of urban environment.