• Journal of Intelligence and Information Systems
• /
• v.24 no.1
• /
• pp.75-100
• /
• 2018

Mobile devices have become an important channel for news content usage in our daily life. However, online news content readers' resistance to online news monetization is more serious than other digital content businesses, such as webtoons, music sources, videos, and games. Since major portal sites distribute online news content free of charge to increase their traffics, customers have been accustomed to free news content; hence this makes online news providers more difficult to switch their policies on business models (i.e., monetization policy). As a result, most online news providers are highly dependent on the advertising business model, which can lead to increasing number of false, exaggerated, or sensational advertisements inside the news website to maximize their advertising revenue. To reduce this advertising dependencies, many online news providers had attempted to switch their 'free' readers to 'paid' users, but most of them failed. However, recently, some online news media have been successfully applying the Pay-What-You-Want (PWYW) payment model, which allows readers to voluntarily pay fees for their favorite news content. These successful cases shed some lights to the managers of online news content provider regarding that the PWYW model can serve as an alternative business model. In this study, therefore, we collected 379 online news articles from Ohmynews.com that has been successfully employing the PWYW model, and analyzed the comparative importance of systematic attributes of online news content on readers' voluntary payment. More specifically, we derived the six systematic attributes (i.e., Type of Article Title, Image Stimulation, Article Readability, Article Type, Dominant Emotion, and Article-Image Similarity) and three or four levels within each attribute based on previous studies. Then, we conducted content analysis to measure five attributes except Article Readability attribute, measured by Flesch readability score. Before conducting main content analysis, the face reliabilities of chosen attributes were measured by three doctoral level researchers with 37 sample articles, and inter-coder reliabilities of the three coders were verified. Then, the main content analysis was conducted for two months from March 2017 with 379 online news articles. All 379 articles were reviewed by the same three coders, and 65 articles that showed inconsistency among coders were excluded before employing conjoint analysis. Finally, we examined the comparative importance of those six systematic attributes (Study 1), and levels within each of the six attributes (Study 2) through conjoint analysis with 314 online news articles. From the results of conjoint analysis, we found that Article Readability, Article-Image Similarity, and Type of Article Title are the most significant factors affecting online news readers' voluntary payment. First, it can be interpreted that if the level of readability of an online news article is in line with the readers' level of readership, the readers will voluntarily pay more. Second, the similarity between the content of the article and the image within it enables the readers to increase the information acceptance and to transmit the message of the article more effectively. Third, readers expect that the article title would reveal the content of the article, and the expectation influences the understanding and satisfaction of the article. Therefore, it is necessary to write an article with an appropriate readability level, and use images and title well matched with the content to make readers voluntarily pay more. We also examined the comparative importance of levels within each attribute in more details. Based on findings of two studies, two major and nine minor propositions are suggested for future empirical research. This study has academic implications in that it is one of the first studies applying both content analysis and conjoint analysis together to examine readers' voluntary payment behavior, rather than their intention to pay. In addition, online news content creators, providers, and managers could find some practical insights from this research in terms of how they should produce news content to make readers voluntarily pay more for their online news content.

• Journal of Intelligence and Information Systems
• /
• v.26 no.1
• /
• pp.135-149
• /
• 2020

Artificial intelligences are changing world. Financial market is also not an exception. Robo-Advisor is actively being developed, making up the weakness of traditional asset allocation methods and replacing the parts that are difficult for the traditional methods. It makes automated investment decisions with artificial intelligence algorithms and is used with various asset allocation models such as mean-variance model, Black-Litterman model and risk parity model. Risk parity model is a typical risk-based asset allocation model which is focused on the volatility of assets. It avoids investment risk structurally. So it has stability in the management of large size fund and it has been widely used in financial field. XGBoost model is a parallel tree-boosting method. It is an optimized gradient boosting model designed to be highly efficient and flexible. It not only makes billions of examples in limited memory environments but is also very fast to learn compared to traditional boosting methods. It is frequently used in various fields of data analysis and has a lot of advantages. So in this study, we propose a new asset allocation model that combines risk parity model and XGBoost machine learning model. This model uses XGBoost to predict the risk of assets and applies the predictive risk to the process of covariance estimation. There are estimated errors between the estimation period and the actual investment period because the optimized asset allocation model estimates the proportion of investments based on historical data. these estimated errors adversely affect the optimized portfolio performance. This study aims to improve the stability and portfolio performance of the model by predicting the volatility of the next investment period and reducing estimated errors of optimized asset allocation model. As a result, it narrows the gap between theory and practice and proposes a more advanced asset allocation model. In this study, we used the Korean stock market price data for a total of 17 years from 2003 to 2019 for the empirical test of the suggested model. The data sets are specifically composed of energy, finance, IT, industrial, material, telecommunication, utility, consumer, health care and staple sectors. We accumulated the value of prediction using moving-window method by 1,000 in-sample and 20 out-of-sample, so we produced a total of 154 rebalancing back-testing results. We analyzed portfolio performance in terms of cumulative rate of return and got a lot of sample data because of long period results. Comparing with traditional risk parity model, this experiment recorded improvements in both cumulative yield and reduction of estimated errors. The total cumulative return is 45.748%, about 5% higher than that of risk parity model and also the estimated errors are reduced in 9 out of 10 industry sectors. The reduction of estimated errors increases stability of the model and makes it easy to apply in practical investment. The results of the experiment showed improvement of portfolio performance by reducing the estimated errors of the optimized asset allocation model. Many financial models and asset allocation models are limited in practical investment because of the most fundamental question of whether the past characteristics of assets will continue into the future in the changing financial market. However, this study not only takes advantage of traditional asset allocation models, but also supplements the limitations of traditional methods and increases stability by predicting the risks of assets with the latest algorithm. There are various studies on parametric estimation methods to reduce the estimated errors in the portfolio optimization. We also suggested a new method to reduce estimated errors in optimized asset allocation model using machine learning. So this study is meaningful in that it proposes an advanced artificial intelligence asset allocation model for the fast-developing financial markets.

• Journal of Intelligence and Information Systems
• /
• v.26 no.1
• /
• pp.1-21
• /
• 2020

With the rapid development of artificial intelligence technology, various techniques have been developed to extract meaningful information from unstructured text data which constitutes a large portion of big data. Over the past decades, text mining technologies have been utilized in various industries for practical applications. In the field of business intelligence, it has been employed to discover new market and/or technology opportunities and support rational decision making of business participants. The market information such as market size, market growth rate, and market share is essential for setting companies' business strategies. There has been a continuous demand in various fields for specific product level-market information. However, the information has been generally provided at industry level or broad categories based on classification standards, making it difficult to obtain specific and proper information. In this regard, we propose a new methodology that can estimate the market sizes of product groups at more detailed levels than that of previously offered. We applied Word2Vec algorithm, a neural network based semantic word embedding model, to enable automatic market size estimation from individual companies' product information in a bottom-up manner. The overall process is as follows: First, the data related to product information is collected, refined, and restructured into suitable form for applying Word2Vec model. Next, the preprocessed data is embedded into vector space by Word2Vec and then the product groups are derived by extracting similar products names based on cosine similarity calculation. Finally, the sales data on the extracted products is summated to estimate the market size of the product groups. As an experimental data, text data of product names from Statistics Korea's microdata (345,103 cases) were mapped in multidimensional vector space by Word2Vec training. We performed parameters optimization for training and then applied vector dimension of 300 and window size of 15 as optimized parameters for further experiments. We employed index words of Korean Standard Industry Classification (KSIC) as a product name dataset to more efficiently cluster product groups. The product names which are similar to KSIC indexes were extracted based on cosine similarity. The market size of extracted products as one product category was calculated from individual companies' sales data. The market sizes of 11,654 specific product lines were automatically estimated by the proposed model. For the performance verification, the results were compared with actual market size of some items. The Pearson's correlation coefficient was 0.513. Our approach has several advantages differing from the previous studies. First, text mining and machine learning techniques were applied for the first time on market size estimation, overcoming the limitations of traditional sampling based- or multiple assumption required-methods. In addition, the level of market category can be easily and efficiently adjusted according to the purpose of information use by changing cosine similarity threshold. Furthermore, it has a high potential of practical applications since it can resolve unmet needs for detailed market size information in public and private sectors. Specifically, it can be utilized in technology evaluation and technology commercialization support program conducted by governmental institutions, as well as business strategies consulting and market analysis report publishing by private firms. The limitation of our study is that the presented model needs to be improved in terms of accuracy and reliability. The semantic-based word embedding module can be advanced by giving a proper order in the preprocessed dataset or by combining another algorithm such as Jaccard similarity with Word2Vec. Also, the methods of product group clustering can be changed to other types of unsupervised machine learning algorithm. Our group is currently working on subsequent studies and we expect that it can further improve the performance of the conceptually proposed basic model in this study.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.40 no.5
• /
• pp.100-108
• /
• 2012

The purpose of the study was to evaluate the influence of shading and ventilation on Mean Radiant Temperature(MRT) of the outdoor space at a summer outdoor. The Wind Speed(WS), Air Temperature(AT) and Globe Temperature(GT) were recorded every minute from $1^{st}$ of May to the $30^{th}$ of September 2011 at a height of 1.2m above in four experimental plots with different shading and ventilating conditions, with a measuring system consisting of a vane type anemometer(Barini Design's BDTH), Resistance Temperature Detector(RTD, Pt-100), standard black globe(${\O}$ 150mm) and data acquisition systems(National Instrument's Labview and Compfile Techs' Moacon). To implement four different ventilating and shading conditions, three hexahedral steel frames, and one natural plot were established in the open grass field. Two of the steel frames had a dimension of $3m(W){\times}3m(L){\times}1.5m(H)$ and every vertical side covered with transparent polyethylene film to prevent lateral ventilation(Ventilation Blocking Plot: VP), and an additional shading curtain was applied on the top side of a frame(Shading and Ventilation Blocking Plot: SVP). The third was $1.5m(W){\times}1.5m(L){\times}1.5m(H)$, only the top side of which was covered by the shading curtain without the lateral film(Shading Plot: SP). The last plot was natural condition without any kind of shading and wind blocking material(Natural Open Plot: NP). Based on the 13,262 records of 44 sunny days, the time serial difference of AT and GT for 24 hour were analyzed and compared, and statistical analysis was done based on the 7,172 records of daytime period from 7 A.M. to 8 P.M., while the relation between the MRT and solar radiation and wind speed was analyzed based on the records of the hottest period from 11 A.M. to 4 P.M.. The major findings were as follows: 1. The peak AT was $40.8^{\circ}C$ at VP and $35.6^{\circ}C$ at SP showing the difference about $5^{\circ}C$, but the difference of average AT was very small within${\pm}1^{\circ}C$. 2. The difference of the peak GT was $12^{\circ}C$ showing $52.5^{\circ}C$ at VP and $40.6^{\circ}C$ at SP, while the gap of average GT between the two plots was $6^{\circ}C$. Comparing all four plots including NP and SVP, it can be said that the shading decrease $6^{\circ}C$ GT while the wind blocking increase $3^{\circ}C$ GT. 3. According to the calculated MRT, the shading has a cooling effect in reducing a maximum of $13^{\circ}C$ and average $9^{\circ}C$ MRT, while the wind blocking has heating effect of increasing average $3^{\circ}C$ MRT. In other words, the MRT of the shaded area with natural ventilation could be cooler than the wind blocking the sunny site to about $16^{\circ}C$ MRT maximum. 4. The regression and correlation tests showed that the shading is more important than the ventilation in reducing the MRT, while both of them do an important role in improving the outdoor thermal comfort. In summary, the results of this study showed that the shade is the first and the ventilation is the second important factor in terms of improving outdoor thermal comfort in summer daylight hours. Therefore, it can be apparently said that the more shade by the forest, shading trees etc., the more effective in conditioning the microclimate of an outdoor space reducing the useless or even harmful heat energy for human activities. Furthermore, the delicately designed wind corridor or outdoor ventilation system can improve even the thermal environment of urban area.

• Journal of Intelligence and Information Systems
• /
• v.24 no.2
• /
• pp.221-241
• /
• 2018

Deep learning is getting attention recently. The deep learning technique which had been applied in competitions of the International Conference on Image Recognition Technology(ILSVR) and AlphaGo is Convolution Neural Network(CNN). CNN is characterized in that the input image is divided into small sections to recognize the partial features and combine them to recognize as a whole. Deep learning technologies are expected to bring a lot of changes in our lives, but until now, its applications have been limited to image recognition and natural language processing. The use of deep learning techniques for business problems is still an early research stage. If their performance is proved, they can be applied to traditional business problems such as future marketing response prediction, fraud transaction detection, bankruptcy prediction, and so on. So, it is a very meaningful experiment to diagnose the possibility of solving business problems using deep learning technologies based on the case of online shopping companies which have big data, are relatively easy to identify customer behavior and has high utilization values. Especially, in online shopping companies, the competition environment is rapidly changing and becoming more intense. Therefore, analysis of customer behavior for maximizing profit is becoming more and more important for online shopping companies. In this study, we propose 'CNN model of Heterogeneous Information Integration' using CNN as a way to improve the predictive power of customer behavior in online shopping enterprises. In order to propose a model that optimizes the performance, which is a model that learns from the convolution neural network of the multi-layer perceptron structure by combining structured and unstructured information, this model uses 'heterogeneous information integration', 'unstructured information vector conversion', 'multi-layer perceptron design', and evaluate the performance of each architecture, and confirm the proposed model based on the results. In addition, the target variables for predicting customer behavior are defined as six binary classification problems: re-purchaser, churn, frequent shopper, frequent refund shopper, high amount shopper, high discount shopper. In order to verify the usefulness of the proposed model, we conducted experiments using actual data of domestic specific online shopping company. This experiment uses actual transactions, customers, and VOC data of specific online shopping company in Korea. Data extraction criteria are defined for 47,947 customers who registered at least one VOC in January 2011 (1 month). The customer profiles of these customers, as well as a total of 19 months of trading data from September 2010 to March 2012, and VOCs posted for a month are used. The experiment of this study is divided into two stages. In the first step, we evaluate three architectures that affect the performance of the proposed model and select optimal parameters. We evaluate the performance with the proposed model. Experimental results show that the proposed model, which combines both structured and unstructured information, is superior compared to NBC(Naïve Bayes classification), SVM(Support vector machine), and ANN(Artificial neural network). Therefore, it is significant that the use of unstructured information contributes to predict customer behavior, and that CNN can be applied to solve business problems as well as image recognition and natural language processing problems. It can be confirmed through experiments that CNN is more effective in understanding and interpreting the meaning of context in text VOC data. And it is significant that the empirical research based on the actual data of the e-commerce company can extract very meaningful information from the VOC data written in the text format directly by the customer in the prediction of the customer behavior. Finally, through various experiments, it is possible to say that the proposed model provides useful information for the future research related to the parameter selection and its performance.

• Tuberculosis and Respiratory Diseases
• /
• v.42 no.2
• /
• pp.206-217
• /
• 1995

Background: In patients with obstructive sleep apnea syndrome(OSAS), there are several factors increasing upper airway resistance and there is a predisposition to compromised respiratory function during waking and sleep related to constitutional factors including a tendency to obesity. Several recent studies have suggested a possible relationship between sleep apnea(SA) and systemic hypertension. But the possible pathophysiologic link between SA and hypertension is still unclear. In this study, we have examined the relationship among age, body mass index(BMI), pulmonary function parameters and polysomnographic data in patients with OSAS. And also we tried to know the difference among these parameters between hypertensive OSAS and normotensive OSAS patients. Methods: Patients underwent a full night of polysomnography and measured pulmonary function during waking. OSAS was diagnosed if patients had more than 5 apneas per hour(apnea index, AI). A careful history of previously known or present hypertension was obtained from each patient, and patients with systolic blood pressure $\geq$ 160mmHg and/or diastolic blood pressure $\geq$ 95mmHg were classified as hypertensives. Results: The noctural nadir of arterial oxygen saturation($SaO_2$ nadir) was negatively related to AI and respiratory disturbance index(RDI), and the degree of noctural oxygen desaturation(DOD) was positively related to AI and RDI. BMI contributed to AI, RDI, $SaO_2$ nadir and DOD values. And also BMI contributed to $FEV_1,\;FEV_1/FVC$ and DLco values. There was a correlation between airway resistance(Raw) and AI, and there was a inverse correlation between DLco and DOD. But there was no difference among these parameters between hypertensive OSAS and normotensive OSAS patients. Conclusion: The obesity contributed to the compromised respiratory function and the severity of OSAS. AI and RDI were important factors in the severity of hypoxia during sleep. The measurement of pulmonary function parameters including Raw and DLco may be helpful in the prediction and assessment of OSAS patients. But we could not find clear difference between hypertensive and normotensive OSAS patients.

• Progress in Medical Physics
• /
• v.27 no.2
• /
• pp.64-71
• /
• 2016

We develop a manufacture procedure for the production of a patient specific customized bolus (PSCB) using a 3D printer (3DP). The dosimetric accuracy of the 3D-PSCB is evaluated for electron beam therapy. In order to cover the required planning target volume (PTV), we select the proper electron beam energy and the field size through initial dose calculation using a treatment planning system. The PSCB is delineated based on the initial dose distribution. The dose calculation is repeated after applying the PSCB. We iteratively fine-tune the PSCB shape until the plan quality is sufficient to meet the required clinical criteria. Then the contour data of the PSCB is transferred to an in-house conversion software through the DICOMRT protocol. This contour data is converted into the 3DP data format, STereoLithography data format and then printed using a 3DP. Two virtual patients, having concave and convex shapes, were generated with a virtual PTV and an organ at risk (OAR). Then, two corresponding electron treatment plans with and without a PSCB were generated to evaluate the dosimetric effect of the PSCB. The dosimetric characteristics and dose volume histograms for the PTV and OAR are compared in both plans. Film dosimetry is performed to verify the dosimetric accuracy of the 3D-PSCB. The calculated planar dose distribution is compared to that measured using film dosimetry taken from the beam central axis. We compare the percent depth dose curve and gamma analysis (the dose difference is 3%, and the distance to agreement is 3 mm) results. No significant difference in the PTV dose is observed in the plan with the PSCB compared to that without the PSCB. The maximum, minimum, and mean doses of the OAR in the plan with the PSCB were significantly reduced by 9.7%, 36.6%, and 28.3%, respectively, compared to those in the plan without the PSCB. By applying the PSCB, the OAR volumes receiving 90% and 80% of the prescribed dose were reduced from $14.40cm^3$ to $0.1cm^3$ and from $42.6cm^3$ to $3.7cm^3$, respectively, in comparison to that without using the PSCB. The gamma pass rates of the concave and convex plans were 95% and 98%, respectively. A new procedure of the fabrication of a PSCB is developed using a 3DP. We confirm the usefulness and dosimetric accuracy of the 3D-PSCB for the clinical use. Thus, rapidly advancing 3DP technology is able to ease and expand clinical implementation of the PSCB.

• The Korean Journal of Food And Nutrition
• /
• v.29 no.6
• /
• pp.835-848
• /
• 2016

The purpose of this study was to find efficient and customized tools for delivering the benefit of health functional foods (HFFs). Delivery tools which could influence the impact of advertising were images, explanations of ingredients, diagrams of health benefit, patents, and comments from authority. Six advertisements were developed using these tools: "A": relevant image + explanation of ingredients + scientific diagram of efficacy; "B": relevant image + explanation of ingredients; "C": relevant image; "D": irrelevant image; "E": irrelevant image + explanation of ingredient + patent; "F": irrelevant image + explanation of ingredient + comments from authority. To analyze the consumer perceptions on HFFs and advertisement effects, 300 respondents were requested to answer a questionnaire comprising of the following questions: 5 questions of attitudes (necessity of HFFs, trust in HFFs, gathering information, watching advertisements and trust in advertisement claims) and 6 questions on the 6 developed advertisements (attention, understanding, sufficiency of information, sympathy, trust, and purchase). Scoring was done as per the 5 Likert scale. There was a higher proportion of females and the elderly, as compared to males and youngsters. The overall consumer attitudes were positive. Explanation of ingredients, scientific diagram of health benefit, patents and expert comments were helpful factors in increasing the advertisement evaluation by consumer, but the images were not. Advertisement evaluation of consumer did not differ with gender and age. However, differences were observed between some of the consumer attitudes (necessity of HFFs, trust in HFFs, gathering information and trust in advertisements claim) and advertisement evaluations (attention, understanding, sympathy and purchase). Our results indicate that for consumers utilizing the HFFs, advertisements with concrete tools such as diagrams, patent, and expert comments are more helpful. However, for consumers who do not have interest in HFFs, the scientific information was irrelevant. We believe that to maximize the effect of health information in advertisements, consumers should be segmented, and customized tools for each segment needs to be developed.

• Radiation Oncology Journal
• /
• v.19 no.4
• /
• pp.327-334
• /
• 2001

Purpose : To find out the role of postoperative adjuvant radiotherapy in the treatment of rectal cancer by comparing survival, pelvic control, complication rate, and any prognostic factor between surgery alone and postoperative radiotherapy group. Materials and methods : From Feb. 1982 to Dec. 1996 total 212 patients were treated by radical surgery with or without postoperative radiotherapy due to rectal carcinoma of modified Astler-Coiler stage $B2\~C3$. Of them, 18 patients had incomplete radiotherapy and so the remaining 194 patients were the database analyzed in this study. One hundred four patients received postoperative radiotherapy and the other 90 patients had surgery only. Radiotherapy was peformed in the range of $39.6\~55.8\;Gy$ (mean: 49.9 Gy) to the whole pelvis and if necessary, tumor bed was boosted by $5.4\~10\;Gy$. Both survival and pelvic control rates were calculated by Kaplan-Meier method and their statistical significance was tested by Log-rank test. Multivariate analysis was peformed by Cox proportional hazards model. Results : 5-year actuarial survival rate (5YSR) and 5-year disease-free survival rate (5YDFSR) of entire patients were $53\%\;and\;49\%$, respectively. 5YSRs of surgery alone group and adjuvant radiotherapy group were $63\%\;vs\;45\%$, respectively (p=0.03). This difference is thought to reflect uneven distribution of stages between two treatment groups (p<0.05 by $\chi^2-test$) with more advanced disease patients in adjuvant radiotherapy group. 5YSRs of surgery alone vs adjuvant radiotherapy group in MAC B2+3, C1, C2+3 were $68\%\;vs\;55\%$ (p=0.09), $100\%\;vs\;100\%$, $40\%\;vs\;33\%$ (p=0.71), respectively. 5YDFSRs of surgery alone vs adjuvant radiotherapy group in above three stages were $65\%\;vs\;49\%$ (p=0.14), $100\%\;vs\;100\%$, $33\%\;vs\;31\%$ (p=0.46), respectively. 5-year pelvic control rate (5YPCR) of entire patients was $72.5\%$. 5YPCRs of surgery alone and adjuvant radiotherapy group were $71\%\;vs\;74\%$, respectively (p=0.41). 5YPCRs of surgery alone vs adjuvant radiotherapy group in B2+3, C1, C2+3 were $79\%\;vs\;75\%$ (p=0.88), $100\%\;vs\;100\%$, $44\%\;vs\;68\%$ (p=0.01), respectively. Multivariate analysis showed that only stage was significant factor affecting overall and disease-free survival in entire patients and also in both treatment groups. In view of pelvic control, stage and operation type were significant in entire patients and only stage in surgery alone group but in adjuvant radiotherapy group, operation type instead of stage was the only significant factor in multivariate analysis as a negative prognostic factor in abdominoperineal resection cases. Conclusion : Our retrospective study showed that postoperative adjuvant radiotherapy could improve the pelvic control in MAC C2+3 group. To improve both pelvic control and survival in all patients with MAC B2 or more, other treatment modality such as concurrent continuous infusion of 5-FU, which is the most standard agent, with radiotherapy should be considered.

• Radiation Oncology Journal
• /
• v.15 no.2
• /
• pp.137-143
• /
• 1997

Purpose : Radiation pneumonitis is one of the complications caused by radiation therapy that includes a Portion of the lung tissue. The severity of radiation induced pulmonary dysfunction depends on the irradiated lung volume, total dose, dose rate and underlying Pulmonary function. It also depends on whether chemotherapy is done or not. The irradiated lung volume is the most important factor to predict the pulmonary dysfunction in breast cancer Patients following radiation therapy. There are some data that show the irradiated lung volume measured from CT scans as a part of treatment Planning with the tangential beams. But such data have not been reported in Korea. We planned to evaluate the irradiated lung volume quantitatively using CT scans for the breast tangential field and search for useful factors that could Predict the irradiated lung volume Materials and Methods : The lung volume was measured for 25 patients with breast cancer irradiated with tangential field from Jan.1995 to Aug.1996. Parameters that can predict the irradiated lung volume included; (1) the peruendicular distance from the Posterior tangential edge to the posterior part of the anterior chest wall at the center of the field (CLD) ; (2) the maximum perpendicular distance from the posterior tangential field edge to the posterior Part of the anterior chest wall (MLD) ; (3) the greatest perpendicular distance from the Posterior tangential edge to the posterior part of anterior chest wall on CT image at the center of the longitudinal field (GPD) ; (4) the length of the longitudinal field (L). The irradiated lung volume(RV), the entire both lung volume(EV) and the ipsilateral lung volume(IV) were measured using dose volume histogram. The relationship between the irradiated lung volume and predictors was evaluated by regression analysis. Results :The RV is 61-279cc (mean 170cc), the RV/EV is $2.9-13.0\%\;(mean\;5.8\%)$ and the RV/IV is $4.9-29.0\%\;(mean\;12.2\%)$. The CLD, the MLD and the GPD ave 1.9-3.3cm, 1.9-3.3cm and 1.4-3.1cm respectively. The significant relations between the irradiated lung volume such as RV. RV/EV, RV/IV and parameters such as CLD, MLD, GPO, L. $CLD\timesL,\;MLD\timesL\;and\;GPD\timesL$ are not found with little variances in parameters. The RV/IV of the left breast irradiation is significantly larger than that of the right but the RV/EVS do not show the differences. There is no symptomatic radiation pneumonitis at least during 6 months follow up. Conclusion : The significant relationship between the irradiated lung volume and predictors is not found with little variation on parameters. The irradiated lung volume in the tangential held is liss than $10\%$ of entire lung volume when CLO is less than 3cm. The RV/IV of the left tangential field is larger than that of the right but there was no significant differences in RV/EVS. Symptomatic radiation pneumonitis has not occurred during minimum 6 months follow up.