• Journal of Intelligence and Information Systems
• /
• v.23 no.4
• /
• pp.77-110
• /
• 2017

Recently, the diversification and individualization of consumption patterns through the web and mobile devices based on the Internet have been rapid. As this happens, the efficient operation of the offline store, which is a traditional distribution channel, has become more important. In order to raise both the sales and profits of stores, stores need to supply and sell the most attractive products to consumers in a timely manner. However, there is a lack of research on which SKUs, out of many products, can increase sales probability and reduce inventory costs. In particular, if a company sells products through multiple in-store stores across multiple locations, it would be helpful to increase sales and profitability of stores if SKUs appealing to customers are recommended. In this study, the recommender system (recommender system such as collaborative filtering and hybrid filtering), which has been used for personalization recommendation, is suggested by SKU recommendation method of a store unit of a distribution company that handles a homogeneous brand through a plurality of sales stores by country and region. We calculated the similarity of each store by using the purchase data of each store's handling items, filtering the collaboration according to the sales history of each store by each SKU, and finally recommending the individual SKU to the store. In addition, the store is classified into four clusters through PCA (Principal Component Analysis) and cluster analysis (Clustering) using the store profile data. The recommendation system is implemented by the hybrid filtering method that applies the collaborative filtering in each cluster and measured the performance of both methods based on actual sales data. Most of the existing recommendation systems have been studied by recommending items such as movies and music to the users. In practice, industrial applications have also become popular. In the meantime, there has been little research on recommending SKUs for each store by applying these recommendation systems, which have been mainly dealt with in the field of personalization services, to the store units of distributors handling similar brands. If the recommendation method of the existing recommendation methodology was 'the individual field', this study expanded the scope of the store beyond the individual domain through a plurality of sales stores by country and region and dealt with the store unit of the distribution company handling the same brand SKU while suggesting a recommendation method. In addition, if the existing recommendation system is limited to online, it is recommended to apply the data mining technique to develop an algorithm suitable for expanding to the store area rather than expanding the utilization range offline and analyzing based on the existing individual. The significance of the results of this study is that the personalization recommendation algorithm is applied to a plurality of sales outlets handling the same brand. A meaningful result is derived and a concrete methodology that can be constructed and used as a system for actual companies is proposed. It is also meaningful that this is the first attempt to expand the research area of the academic field related to the existing recommendation system, which was focused on the personalization domain, to a sales store of a company handling the same brand. From 05 to 03 in 2014, the number of stores' sales volume of the top 100 SKUs are limited to 52 SKUs by collaborative filtering and the hybrid filtering method SKU recommended. We compared the performance of the two recommendation methods by totaling the sales results. The reason for comparing the two recommendation methods is that the recommendation method of this study is defined as the reference model in which offline collaborative filtering is applied to demonstrate higher performance than the existing recommendation method. The results of this model are compared with the Hybrid filtering method, which is a model that reflects the characteristics of the offline store view. The proposed method showed a higher performance than the existing recommendation method. The proposed method was proved by using actual sales data of large Korean apparel companies. In this study, we propose a method to extend the recommendation system of the individual level to the group level and to efficiently approach it. In addition to the theoretical framework, which is of great value.

• The Journal of Korean Society for Radiation Therapy
• /
• v.21 no.1
• /
• pp.1-7
• /
• 2009

Purpose: The purpose is to evaluate the accuracy of correcting the targeting error through the Target Location System (TLS) for the location change error of the reference point which arises from the movement or motion of patient during the treatment using the CyberKnife. Materials and Methods: In this test, Gafchromic MD-55 film was inserted into the head and neck phantom to analyze the accuracy of the targeting, and then the 6 MV X-ray of CyberKnife (CyberKnife Robotic Radiosurgery System G4, Accuray, US) was irradiated. End to End (E2E) program was used to analyze the accuracy of targeting, which is provided by Accuray Corporation. To compute the error of the targeting, the test was carried out with the films that were irradiated 12 times by maintaining the distance within the rage of $0{\pm}0.2\;mm$ toward x, y, z from the reference point and maintaining the angle within the rage of $0{\pm}0.2^{\circ}$ toward roll, pitch, yaw, and then with the films which were irradiated 6 times by applying intentional movement. And the correlation in the average value of the reference film and the test film were analyzed through independent samples t-test. In addition, the consistency of dose distribution through gamma-index method (dose difference: 3%) was quantified, compared, and analyzed by varying the distance to agreement (DTA) to 1 mm, 1.5 mm, 2 mm, respectively. Results: E2E test result indicated that the average error of the reference film was 0.405 mm and the standard deviation was 0.069 mm. The average error of the test film was 0.413 mm with the standard deviation of 0.121 mm. The result of independent sampling t-test for both averages showed that the significant probability was P=0.836 (confidence level: 95%). Besides, by comparing the consistency of dose distribution of DTA through 1 mm, 1.5 mm, 2 mm, it was found that the average dose distribution of axial film was 95.04%, 97.56%, 98.13%, respectively in 3,314 locations of the reference film, consistent with the average dose distribution of sagittal film that was 95.47%, 97.68%, 98.47%, respectively. By comparing with the test film, it was found that the average dose distribution of axial film was 96.38%, 97.57%, 98.04%, respectively, at 3,323 locations, consistent with the average dose distribution of sagittal film which was 95.50%, 97.87%, 98.36%, respectively. Conclusion: Robotic CyberKnife traces and complements in real time the error in the location change of the reference point caused by the motion or movement of patient during the treatment and provides the accuracy with the consistency of over 95% dose distribution and the targeting error below 1 mm.

• The KIPS Transactions:PartB
• /
• v.11B no.6
• /
• pp.749-758
• /
• 2004

In this paper, we propose a new method utilizing only raw corpus without additional human effort for disambiguation of target word selection in English-Korean machine translation. We use two data-driven techniques; one is the Latent Semantic Analysis(LSA) and the other the Probabilistic Latent Semantic Analysis(PLSA). These two techniques can represent complex semantic structures in given contexts like text passages. We construct linguistic semantic knowledge by using the two techniques and use the knowledge for target word selection in English-Korean machine translation. For target word selection, we utilize a grammatical relationship stored in a dictionary. We use k- nearest neighbor learning algorithm for the resolution of data sparseness Problem in target word selection and estimate the distance between instances based on these models. In experiments, we use TREC data of AP news for construction of latent semantic space and Wail Street Journal corpus for evaluation of target word selection. Through the Latent Semantic Analysis methods, the accuracy of target word selection has improved over 10% and PLSA has showed better accuracy than LSA method. finally we have showed the relatedness between the accuracy and two important factors ; one is dimensionality of latent space and k value of k-NT learning by using correlation calculation.

• The Journal of Korean Society for Radiation Therapy
• /
• v.29 no.2
• /
• pp.19-26
• /
• 2017

Objectives: In the Lung, the VMAT rotates continuously and examines radiation. That increases the low doses to normal lung. Due to that, the incidence of radiation pneumonia among radiation side effects may increase. The cause of radiation pneumonia is the lower dose area of the lungs. The H-VMAT was applied to patients who applied to reduce radiation in the lower doses of the lungs. We wanted to assess the usefulness of the H-VMAT by comparing the radiation doses to the low dose areas of the lungs and the normal organs. Materials and Methods: A total of 26 patients who applied for a H-VMAT procedure were applied to the patient. The prescription dose applied to total dose 44 Gy from 22 divisions. For each patient, a plan was implemented with Conventional RT, VMAT and H-VMAT. Conventional RT was carried out in four to five fields each, considering the size, location, shape, and location of the PTV. In the case of a VMAT plan, the two Half ARC, three Half ARC method and the two Full ARC were planned. The H-VMAT was planned by adding two Static fields in the VMAT, taking into account the dose of the lung and the tolerance dose of the organs. Results: In the NSCLC, the lung doses $V_5$ and $V_{10}$ of the lungs except for the treatment plan volume were the lowest with $55.40{\pm}13.39%$ and $32.05{\pm}11.37%$ of H-VMAT. And, in the SCLC, the lung doses of V5 and V10 were the lowest at $64.32{\pm}16.15%$ and $35.50{\pm}9.91%$, respectively. The spinal dose of VMAT in NSCLC was $21.15{\pm}4.02Gy$, which was 7.94 Gy lower than other treatment methods. The lowest spinal dose was delivered at $19.72{\pm}1.82Gy$ for SCLC. The mean dose delivered to the esophagus was also $17.44{\pm}2.04Gy$ and $17.84{\pm}9.20Gy$ in SCLC and NSCLC, respectively. Conclusion: When comparing the value of the surrounding normal organ dose, the VMAT showed that less doses were transmitted from the heart, esophagus and spinal cord than the rest of the treatment plan. However, it was similar to VMAT in normal organs except for the spinal cord. VMAT has increased doses of some normal organs but did not exceed the tolerance dose. It showed a low value in $V_5$, $V_{10}$. When comparing Conventional RT, VMAT, and H-VMAT, If the dose to the heart, esophagus and spinal cord is lower than the tolerance dose, it is thought to reduce the incidence of radiation pneumonia by applying H-VMAT that show the benefits of low doses of the lungs.

• Radiation Oncology Journal
• /
• v.25 no.1
• /
• pp.43-53
• /
• 2007

[ $\underline{Purpose}$ ]: To evaluate the relationship between the expression of EGFR, p53, Cox-2, Bcl-2 and the clinical parameters of NPC (nasopharyngeal carcinoma) patients treated with radiotherapy with/without chemotherapy, and to determine if these could be used as a biologic marker. $\underline{Materials\;and\;Methods}$: This study retrospectively examined 75 NPC patients who were pathologically diagnosed at St. Mary's Hospital and Kangnam St Mary's Hospital from March 1988 to August 2002 and treated with radiotherapy with/without chemotherapy. The levels of EGFR, p53, Cox-2, and Bcl-2 expression were determined immunohistochemically. The relationship between the levels of EGFR, p53, Cox-2 and Bcl-2 expression and the H- E staining findings including the WHO classification, TNM stage, tumor response to chemotherapy and radiotherapy, disease free survival (DFS), and overall survival (OS) was analyzed. $\underline{Results}$: At a median follow up of 50.8 months (range: $5.5{\sim}201$ months), the 3 years OS rate and PFS rate were 68.7% and 68.2%, respectively. The five year OS rate and PFS rate were 53.5% and 51.1%, respectively. The median OS duration and PFS duration were 85.5 months and 61.1 months, respectively. The WHO classification correlated with the complete response rate, lymph node metastasis and distant metastasis. The expression of p53 was associated with increased mitosis and poor overall survival. The expression of Bcl-2 correlated with the DFS and WHO classification. The expression of Cox-2 correlated with a poor overall survival and response rate in the lymph node. However, EGFR was not correlated with any factors. $\underline{Conclusion}$: These results suggest that the expression of p53, Cox-2, Bcl-2 plays role in predicting prognostic factors for NPC treated with radiotherapy with/without chemotherapy. However, further study on a larger number of patients will be needed to identify more useful biomarkers of NPC.

• Radiation Oncology Journal
• /
• v.25 no.4
• /
• pp.261-267
• /
• 2007

Purpose: To evaluate the association between radiation pneumonitis and dose-volume histogram parameters and to provide practical guidelines to prevent radiation pneumonitis following radiotherapy administered for breast cancer including internal mammary lymph nodes. Materials and Methods: Twenty patients with early breast cancer who underwent a partial mastectomy were involved in this study. The entire breast, supraclavicular lymph nodes, and internal mammary lymph nodes were irradiated with a dose of 50.4 Gy in 28 fractions. Radiation pneumonitis was assessed by both radiological pulmonary change (RPC) and by evaluation of symptomatic radiation pneumonitis. Dose-volume histogram parameters were compared between patients with grade <2 RPC and those with grade ${\geq}$2 RPC. The parameters were the mean lung dose, V10 (percent lung volume receiving equal to and more than 10 Gy), V20, V30, V40, and normal tissue complication probability (NTCP). Results: Of the 20 patients, 9 (45%) developed grade 2 RPC and 11 (55%) did not develop RPC (grade 0). Only one patient developed grade 1 symptomatic radiation pneumonitis. Univariate analysis showed that among the dose-volume histogram parameters, NTCP was significantly different between the two RPC grade groups (p<0.05). Fisher's exact test indicated that an NTCP value of 45% was appropriate as an RPC threshold level. Conclusion: This study shows that NTCP can be used as a predictor of RPC after radiotherapy of the internal mammary lymph nodes in breast cancer. Clinically, it indicates that an RPC is likely to develop when the NTCP is greater than 45%.

• Tuberculosis and Respiratory Diseases
• /
• v.44 no.6
• /
• pp.1296-1307
• /
• 1997

Background : Erdosteine is a thiol derivative developed for the treatement of chronic obstructive bronchitis, including acute infective exacerbation of chronic bronchitis. Erdosteine has mucomodulating and antioxidant properties and especially exhibits excellent gastrointestinal tolerability. Methods : The study was conducted as a prospective evaluation, with 2 comparative groups orally treated with erdosteine 300mg (bid.) or ambroxol 30mg (b.i.d.) for 7 days and the design of trial was double-blind. The treatments have been assigned randomly to patients (n=80) with acute or chronic bronchitis. The primary end-point used to determine efficacy in this study was subjective symptoms including expectorating frequence, expectoration volume, expectorating difficulty, expectoration viscosity, cough intensity and dyspnea. The secondary end-points of efficacy was the result of arterial blood gas analysis and pulmonary function test. Safety was evaluated with adverse drug reactions and laboratory tests monitoring. 61 patients was included in the efficacy analysis, due to the fact that 19 patients drop-out for different reasons. The obtained values have been analyzed with paired Hest., ANOVA test., multivariate $t^2$-test, repeated measures analysis of covariance, two sample t-test, loglinear-logit model analysis, Fisher's exact test. Results : 1) There was no significant difference on demographic data and vital signs between erdosteine and ambroxol treated groups. 2) The comparison between erdosteine and ambroxol treated groups showed no significant difference in improvement of each symptom in spite of the more favorable efficacy obtained with erdosteine. No difference on the contrary was observed for arterial blood gas analysis and pulmonary function test. 3) As safety is concerned, no clinical significant changes in laboratory test and symptom were induced in erdosteine and ambroxol treated group and two patients in ambroxol treated group drop-out for adverse reactions in symptom. 4) In the evaluation of final clinical efficacy, erdosteine improved more effectively patient's overall symptoms {very good effect (11/31), good effect (12/31), moderate effect (6/31), no effect (2/31), aggravation (0/31)} than ambroxol {very good effect (6/30), good effect (14/30), moderate effect (5/30), no effect (4/30), aggravation (2/30)}. And the probability of symptomatic improvement by erdosteine compared to ambroxol was 2.5 times. (p<0.05). Conclusion : This study showed that erdosteine was clinically effective and safe drug for treatment of acute and chronic bronchitis.