• Korean Journal of Heritage: History & Science
• /
• v.55 no.4
• /
• pp.212-240
• /
• 2022

This study is a multi-layered exploration of 「The Thirty-Six Scenery of Seongdeok Summer Mountain Resort(承德避暑山莊三十六景)」 (The 36th view of Kangxi) recited by Emperor Kangxi of China through literature study, ancient calligraphy diagrams, and field studies. The conclusion of tracing the landscape characteristics and implications contained in 「The 36th view of Kangxi」 through the analysis of the headword(標題語) and the interpretation of the Jeyeong poem(題詠詩) is as follows. 「The 36th view of Kangxi」 is an extension of the outer edge of the Eight Sceneries, and when compared to the existing Eight Sceneries peom and Eight Sceneries painting, it is found that the landscape is centered on the 'viewpoint' rather than the landscape object. In particular, it aimed to create a structured landscape centered on nine types of buildings represented by 'Jeon(殿)' and 'Jeong(亭)' was given. In particular, Yeouiju, located in Lake district, is a scenic country endowed with the character of a gardens in Garden, which is composed by collecting famous representative Chinese landscapes and landscapes of Sansu-si and Sanshu Painting. As a result of headword analysis to understand the characteristics of landscape components, 14 landscapes (38.9%) related to water elements and 13 landscapes(36.1%) related to mountain elements, the elements related to architecture and civil engineering were classified in the order of 3 cases(8.3%), and the elements related to the skylight were classified in the order of 2 cases(5.6%). However, in Jeyeong-si, the mention of landscape vocabulary for climate elements was overwhelming. In other words, in the poems of 「The 36th Scenery of Kangxi」, scenery vocabulary symbolizing 'coolness' such as 雲(cloud), 水(water), 泉(spring), 清(clear), 波(wave), 流(wave), 風(wind) and 無暑(without heat), etc. It is not a coincidence that it appears, and it is strongly attached to the sense of place of Summer Mountain Resort in Rehe(熱河). Among the 23 landscapes whose seasonal background was confirmed, the fact that the lower landscape is portrayed as the majority and the climate elements of the resort area are portrayed in three-dimensional and multi-dimensional ways are closely related to the period of enjoying the gardens of Kangxi, the main subject of the landscape. In addition, many animal and plant landscapes appearing in Jeyeong-si appear to be in the same context as the spatial attributes of not only recreation, but also contemplation and hunting. On the other hand, in Jeyeongsi, there are 33 wonders(91.7%) citing famous people and famous books through ancient poems, old stories, and ancient stories tends to be prominent. It is inferred that this was based on Kangxi's understanding and pride in traditional Chinese culture. In 「The 36th view of Kangxi」, not only a book-writing description of the feelings of being entrusted to the family sutras, but also the spirit of patriotism, love, self-discipline and respect for mother and filial piety are strongly implied. Ultimately, 「The 36th view of Kangxi」 shows the real scene of the resort, as well as the spiritual dimension, in a multi-faceted and three-dimensional way, and the spirit of an emperor based on the dignity of the royal family and the sentiments of a writer it deserves to be called a collection of imperial records that were intended to reveal.

• Journal of the korean academy of Pediatric Dentistry
• /
• v.36 no.2
• /
• pp.253-261
• /
• 2009

The purpose of this study was to present new data on the timing and sequence of permanent teeth emergence and to compare these findings with the results of earlier studies. The dental examinations had been performed to E-elementary school students, who visited the Yonsei University Dental Hospital between 1998 and 2005; 1,307 boys and 1,312 girls with the age ranging from 6 to 12 years old. The followings were concluded. 1. Eruption time of the maxillary permanent teeth is as follows. It was 6.81 years in male and 6.73 years in female for the central incisor, 7.78 years in male and 7.65 years in female for the lateral incisor, 10.48 years in male and 9.92 years in female for the canine, 9.76 years in male and 9.63 years in female for the first premolar, 10.66 years in male and 10.49 years in female for the second premolar, 6.39 years in male and 6.26 years in female for the first permanent molar, and 12.13 years in male and 12.03 years in female for the second permanent molar. 2. Eruption time of the mandibular permanent teeth is as follows. The central incisor could not be determined in this study, but it is assumed to erupt before the age of 6.08. In the mandible, eruption time was 6.78 years in male and 6.65 years in female for the lateral incisor, 9.76 years in male and 9.05 years in female for the canine, 9.82 years in male and 9.59 years in female for the first premolar, 10.67 years in male and 10.52 years in female for the second premolar, 6.22 years in male and 6.12 years in female for the first permanent molar, and 11.58 years in male and 11.14 years in female for the second permanent molar. 3. The eruption sequence is as follows. In the maxilla, the first permanent molar erupted first, followed by the central incisor, the lateral incisor, the first premolar, the canine, the second premolar, and the second permanent molar. In the mandible, the central incisor erupted first, followed by the first permanent molar, the lateral incisor, the canine, the first premolar, the second premolar, and the second permanent molar. 4. Tooth eruption occurred earlier in female compared to male by average of 0.19 year in the maxilla and 0.29 year in the mandible. 5. In both male and female, the hiatus (interval of rest) occurred between the emergence of lateral incisor and first premolar in the maxilla while it was observed between the lateral incisor and canine in the mandible. Male had a hiatus of 1.98 years in the maxilla and 2.90 years in the mandible, while the female's were 1.98 years and 2.40 years, respectively.

• 한국벤처창업학회:학술대회논문집
• /
• 2009.10a
• /
• pp.265-288
• /
• 2009

The study has found that the main reasons affecting to the mature conjugal relations of middle-aged persons are closely related with sede factors brought about after marriage. Comparatively the factors before marriage have more or less weak power on the conjugal relations. They are mature relations wp between husbands and wives, stable and enough incede, and religious activities, which have deep relations with the satisfactory conjugal relations. C. G. Jung divided the whole life span as self-assuredness period in the first half and self-convergence period in the second half. The first is the period when one does his or her best to get external and physical self. On the contrary, the second is the middle-aged period one finds his or her meaning of life in the religious, philosophical, intuitional, and spiritual world, which lead life into harmony and integration. Therefore if one overcomes some psychological crisis related with middle-aged development he or she can enjoy happy senescence(old age). The study has suggested through literature investigation the definition of middle age and the developmental traits of middle age, and the relations between religions and conjugal relations of middle-aged husbands and wives. Futhermore, it has analyzed the theories which religions have close relations with the life satisfaction of middle-aged conjugal relations. In order to give an analysis the influence of the variable of religious activities and religious maturity, with the degree of conjugal satisfaction, 400 middle age are selected as the object of the study whose ages are ranging from 35 years to 60 years, and who reside in Seoul or near Seoul. They were asked to fill out the questionnaires asking about religious activities, religious maturity, and the conjugal satisfaction from March 25th to April 30th, 2009. The results of the survey have been statistically processed and analyzed. First, the higher religious maturity gives positive influence on the general religious activities including public service, human relations, and spiritual stability. That is, this result indicates that the individual, spiritual, and formal religious activities give to a degree influence on the religious maturity. Second, the maturity of religious life resulting from religious activities has a causation with the satisfaction of conjugal life. In more details, religious activities has a positive influence on the satisfaction of conjugal life(T=31.36, p<.001) In more details, religious activities has a positive influence on the religious maturity(T=31.36, p<.001), and religious activities has a positive influence on the satisfaction of conjugal life(T=33.81, p<.001), and the religious maturity has a positive influence on the satisfaction of conjugal life(T=28.64, p<.001) Third, as we analyze the main effects which religious activities and the religious maturity could give influence on the satisfaction of conjugal life, it is found that both religious activities(F=15.95, p<.001) and the religious maturity(F=23.94, p<.001) give a positive influence on the satisfaction of conjugal life. In conclusion, it is sure that religious activities and the religious maturity have a close relations with the satisfaction of conjugal life. Therefore it can be said that religious activities at the protestant religion, buddhism, and catholic religion can give an important influence on the satisfaction of middle-aged conjugal life.

• Journal of Intelligence and Information Systems
• /
• v.23 no.2
• /
• pp.71-88
• /
• 2017

Language models were originally developed for speech recognition and language processing. Using a set of example sentences, a language model predicts the next word or character based on sequential input data. N-gram models have been widely used but this model cannot model the correlation between the input units efficiently since it is a probabilistic model which are based on the frequency of each unit in the training set. Recently, as the deep learning algorithm has been developed, a recurrent neural network (RNN) model and a long short-term memory (LSTM) model have been widely used for the neural language model (Ahn, 2016; Kim et al., 2016; Lee et al., 2016). These models can reflect dependency between the objects that are entered sequentially into the model (Gers and Schmidhuber, 2001; Mikolov et al., 2010; Sundermeyer et al., 2012). In order to learning the neural language model, texts need to be decomposed into words or morphemes. Since, however, a training set of sentences includes a huge number of words or morphemes in general, the size of dictionary is very large and so it increases model complexity. In addition, word-level or morpheme-level models are able to generate vocabularies only which are contained in the training set. Furthermore, with highly morphological languages such as Turkish, Hungarian, Russian, Finnish or Korean, morpheme analyzers have more chance to cause errors in decomposition process (Lankinen et al., 2016). Therefore, this paper proposes a phoneme-level language model for Korean language based on LSTM models. A phoneme such as a vowel or a consonant is the smallest unit that comprises Korean texts. We construct the language model using three or four LSTM layers. Each model was trained using Stochastic Gradient Algorithm and more advanced optimization algorithms such as Adagrad, RMSprop, Adadelta, Adam, Adamax, and Nadam. Simulation study was done with Old Testament texts using a deep learning package Keras based the Theano. After pre-processing the texts, the dataset included 74 of unique characters including vowels, consonants, and punctuation marks. Then we constructed an input vector with 20 consecutive characters and an output with a following 21st character. Finally, total 1,023,411 sets of input-output vectors were included in the dataset and we divided them into training, validation, testsets with proportion 70:15:15. All the simulation were conducted on a system equipped with an Intel Xeon CPU (16 cores) and a NVIDIA GeForce GTX 1080 GPU. We compared the loss function evaluated for the validation set, the perplexity evaluated for the test set, and the time to be taken for training each model. As a result, all the optimization algorithms but the stochastic gradient algorithm showed similar validation loss and perplexity, which are clearly superior to those of the stochastic gradient algorithm. The stochastic gradient algorithm took the longest time to be trained for both 3- and 4-LSTM models. On average, the 4-LSTM layer model took 69% longer training time than the 3-LSTM layer model. However, the validation loss and perplexity were not improved significantly or became even worse for specific conditions. On the other hand, when comparing the automatically generated sentences, the 4-LSTM layer model tended to generate the sentences which are closer to the natural language than the 3-LSTM model. Although there were slight differences in the completeness of the generated sentences between the models, the sentence generation performance was quite satisfactory in any simulation conditions: they generated only legitimate Korean letters and the use of postposition and the conjugation of verbs were almost perfect in the sense of grammar. The results of this study are expected to be widely used for the processing of Korean language in the field of language processing and speech recognition, which are the basis of artificial intelligence systems.

• Journal of agricultural medicine and community health
• /
• v.26 no.2
• /
• pp.15-41
• /
• 2001

Since medical clients and the community they live in are expected to be center of future public health and medical care system, new service programs must be developed with patients focused on in line with widening public access of information and social participation. Patients- focused service shall mean the area- oriented provision of public health service. In this study, health officers working at public health centers, public health sub- centers and medical offices in Jeonbuk- do area were taken for population in order to investigate their attitudes toward and knowledge about the service area assigning system under the public health programs. Findings from the survey to 260 health officers, divided by general category, are as follows : Government officers at public health organizations appeared to have high grade of understanding to the service area assigning system and also great appreciation for the necessity of it. Regarding the timing for the system to be introduced, they support the gradual implementation and, as for the type of service to be provided, they preferred home nursing and treatment of chronic diseases. Highly positive responses were centered on the health classes under the health promotion projects, and as far as health projects for the old are concerned, services for home nursing, for the disabled and for home- alone people are favored most. On the other hand, budgeting, manpower and reorganization are rated as prerequisite to establishment of the service area assigning system. From the viewpoint of system side, the improvement of working conditions is rendered as most urgent, while the information system for establishing the service area assigning system is conceived far from satisfactory. Proper assignment of specialists was noted as mostly important to establish the delivery system for medical service through the service area assigning system by team. As merits of the service area assigning system, it is pointed out that, through the system, health clients can better be managed and the nursing quality will be improved thank to the enhanced specialization. It is also perceived that the district health service is not well prepared to respond to the increased and diversified needs of community people and, furthermore, service programs of health centers have not been fully developed. The most serious problem standing in the way to expansion of health projects is, it is noted, uniformity (formality) of the project. Based on the results of the survey which suggest time has ripen to introduce the service area assigning system, following strategies are proposed to anchor down the system as soon as possible: First, we should introduce the system gradually, starting from the area selected, and in consideration of area specialities, refraining from the hitherto stereotyped way of providing health service. Second, we should seek to properly assign the specialists and improve the working conditions of the assigned officers by securing sufficient budget, since it is a most urgent step to lay foundation for the service area assigning system. Third, best service program should be developed to meet the satisfaction of community people by responding to their needs and solidifying the management of medical clients. Fourth, wide scope of study should further be conducted in order to help this system take roots in the central living of community residents since pilot project on the experimental base attended by specialists only can not win popularity among the masses.

• Journal of Intelligence and Information Systems
• /
• v.27 no.3
• /
• pp.57-73
• /
• 2021

Maintenance and prevention of failure through anomaly detection of ICT infrastructure is becoming important. System monitoring data is multidimensional time series data. When we deal with multidimensional time series data, we have difficulty in considering both characteristics of multidimensional data and characteristics of time series data. When dealing with multidimensional data, correlation between variables should be considered. Existing methods such as probability and linear base, distance base, etc. are degraded due to limitations called the curse of dimensions. In addition, time series data is preprocessed by applying sliding window technique and time series decomposition for self-correlation analysis. These techniques are the cause of increasing the dimension of data, so it is necessary to supplement them. The anomaly detection field is an old research field, and statistical methods and regression analysis were used in the early days. Currently, there are active studies to apply machine learning and artificial neural network technology to this field. Statistically based methods are difficult to apply when data is non-homogeneous, and do not detect local outliers well. The regression analysis method compares the predictive value and the actual value after learning the regression formula based on the parametric statistics and it detects abnormality. Anomaly detection using regression analysis has the disadvantage that the performance is lowered when the model is not solid and the noise or outliers of the data are included. There is a restriction that learning data with noise or outliers should be used. The autoencoder using artificial neural networks is learned to output as similar as possible to input data. It has many advantages compared to existing probability and linear model, cluster analysis, and map learning. It can be applied to data that does not satisfy probability distribution or linear assumption. In addition, it is possible to learn non-mapping without label data for teaching. However, there is a limitation of local outlier identification of multidimensional data in anomaly detection, and there is a problem that the dimension of data is greatly increased due to the characteristics of time series data. In this study, we propose a CMAE (Conditional Multimodal Autoencoder) that enhances the performance of anomaly detection by considering local outliers and time series characteristics. First, we applied Multimodal Autoencoder (MAE) to improve the limitations of local outlier identification of multidimensional data. Multimodals are commonly used to learn different types of inputs, such as voice and image. The different modal shares the bottleneck effect of Autoencoder and it learns correlation. In addition, CAE (Conditional Autoencoder) was used to learn the characteristics of time series data effectively without increasing the dimension of data. In general, conditional input mainly uses category variables, but in this study, time was used as a condition to learn periodicity. The CMAE model proposed in this paper was verified by comparing with the Unimodal Autoencoder (UAE) and Multi-modal Autoencoder (MAE). The restoration performance of Autoencoder for 41 variables was confirmed in the proposed model and the comparison model. The restoration performance is different by variables, and the restoration is normally well operated because the loss value is small for Memory, Disk, and Network modals in all three Autoencoder models. The process modal did not show a significant difference in all three models, and the CPU modal showed excellent performance in CMAE. ROC curve was prepared for the evaluation of anomaly detection performance in the proposed model and the comparison model, and AUC, accuracy, precision, recall, and F1-score were compared. In all indicators, the performance was shown in the order of CMAE, MAE, and AE. Especially, the reproduction rate was 0.9828 for CMAE, which can be confirmed to detect almost most of the abnormalities. The accuracy of the model was also improved and 87.12%, and the F1-score was 0.8883, which is considered to be suitable for anomaly detection. In practical aspect, the proposed model has an additional advantage in addition to performance improvement. The use of techniques such as time series decomposition and sliding windows has the disadvantage of managing unnecessary procedures; and their dimensional increase can cause a decrease in the computational speed in inference.The proposed model has characteristics that are easy to apply to practical tasks such as inference speed and model management.

• Journal of Preventive Medicine and Public Health
• /
• v.15 no.1
• /
• pp.187-196
• /
• 1982

The followings are the results of the study on industrial accidents occurred at 12 factories manufacturing metal products during the period of 2 years from January 1980 to December 1981 in the area of Yong-Dung-Po in Seoul. The results of the study are as follows: 1. The incidence rate of industrial injuries was 45.7 per 1,000 workers of the sample group and the rate of male (54.0) was three times higher than that of female (17.5). 2. In age groups, the highest rate was observed in the group of under 19 years old with 83.5, while the lowest in the group of 40s. 3. It was found that those who had short term of work experience produced a higher rate of injuries, particularly, the group of workers with less than 1 year of experience showed the highest rate of it as 48.1%. 4. In working time, the highest incidence rate occurred 3 and 7 hours after the beginning of their working showing the rate of 6.0 and 6.1 per 1,000 workers, respectively. 5. The highest incidence rate was observed on Monday as 8.4 per 1,000 workers, and it was 18.3% in aspect of the days of a week. 6. In aspect of the months of a year, the highest incidence was observed on July 1,000 workers and the next was on March as 4.8. These figures account for 11.8% of total occurrence in respective month. as 5. 4 per and 10.5% 7. In causes of injuries, the accident caused by power driven machinery showed the highest rate with 37.5%, the second was due to handling without machinery with 17.2%, and the third was due to falling objects with 14.2%, and striking against objects with 10.2%, and so on. 8. By parts of the body affected, the most injuries 84.3% of them occurred on both upper and lower extremities with the rate of 58.8% for the former and 25.5% for the latter. Fingers were most frequently injured with a rate of 40.3%. Comparing the sites of extremities affected, rate of injuries on the right side was 55.0% and 45.0% on the left side. 9. In the nature of injury, laceration and open wound were the highest with 34. 0%, the next was fracture and dislocation with 31. 9%, and sprain was the third with 8.1%. 10. On the duration of treatment, it lasted less than one month in 68.9% of the injured cases, of which 14.5% of the cases were recovered within 2 weeks, and 54.4% of them were treated more than 2 weeks. And the duration of the treatment tended to be prolonged in larger industries. 11. The ratio of insured accidents to uninsured accidents was 1 to 4.7.

• Journal of the Korean Institute of Traditional Landscape Architecture
• /
• v.35 no.3
• /
• pp.13-21
• /
• 2017

The Cho-yeon Pavilion located in the Wangdae village in Samcheong-ri, Songgwang-myeon, Suncheon-si, was transformed into a place of refuge, a shrine, a vacation home, a lecture hall for kings. Based on the change, the current study has explored the periodic changing placeness and the transformation of cultural landscape and has figured out the meaning. The result of this study is as follows. First, "Cho-yeon", named by Yeonjae Song, Byeong-Seon, originated from Tao Te Ching of Lao Tzu. The concept is found not only in the Cho-yeon Pavilion in Suncheon but also in various places, such as, the Cho-yeon-dae in Pocheon, of the Cho-yeon-dae in Gapyeong, of the Cho-yeon-dae of the embankment behind the Gioheon of Changdeok-gung Garden, Cho-Yeon-Mul-Oe old buildings, including Jung(亭), Dae(臺), Gak(閣), of Ockriukag in Yuseong, etc. This shows that taoistic Poongrhu was naturally grafted onto confucian places, which is one of the examples of the fusion of Confucianism, Buddhism, and Taoism. Second, the placeness of the Cho-yeon Pavilion area is related to a legend that King Gong-min sought refuge here at the end of the Koryo Dynasty. The legend is based on the Wangdae village(king's region), Yu-Gyeong(留京)(the place where kings stayed), rock inscription of Wang-Dae-Sa-Jeok, Oh-Jang-Dae (the place where admiral flags were planted), and the Mohusan Mountain. Third, the Cho-yeon Pavilion not only has a base(the vacation home) that reflects confucian values from the rock inscription(趙鎭忠別業, 趙秉翼, 宋秉璿) of the beautiful rock walls and torrents but also has territoriality as taoistic Abode of the Immortals (there are places where people believe taoist hermits with miraculous powers live within 1km of the pavillion: Wol-Cheong(月靑), Pung-Cheong(風靑), Su-Cheong(水靑), Dong-Cheon(洞天). The Cho-yeon Pavilion also reflects the heaven of Neo-Confucianism for, pursuing study, and improving aesthetic sense by expanding its outer area and establishing the nine Gok: Se-Rok-Gyo(洗鹿橋)., Bong-Il-Dae(捧日臺), Ja-Mi-Gu(紫薇鳩), Un-Mae-Dae(雲梅臺), Wa-Ryong-Chong(臥龍叢), Gwang-Seok-Dae(廣石臺), Eun-Seon-Gul(隱仙窟), Byeok-Ok-Dam(碧玉潭), and Wa-Seok-Po(臥石布). In sum, the Cho-yeon Pavilion is a complex cultural landscape. Fourth, the usage of the Cho-yeon Pavilion was expanded and transformed: (1)Buddhist monastery${\rightarrow}$(2)Confucian vacation home${\rightarrow}$(3)Vacation home+Taoistic Poongrhu Place${\rightarrow}$(4)Vacation Home+Taoistic Poongrhu Place+Lecture Hall(the heaven of Neo-Confucianism). To illustrate, in 7978, the place served as Buddist Monk Kwang-Sa's monastery; in 1863, Cho, Jin-Choong established a vacation home by building a shrine in front of the tomb of his ancestor; in 1864, Cho, Jae-Ho expanded its usage to a vacation home to serve ancestors as a taoistic place by repairing the pavilion with roof tiles; and after 1890, Cho, Jun-Sup received the name of the pavilion, Cho-yeon, from his teacher Song, Byeong-Seon, and used the Pavilion for a lecture hall.

• The Korean Journal of Air & Space Law and Policy
• /
• v.23 no.2
• /
• pp.31-66
• /
• 2008

Until Montreal Convention was established in 1999, the Warsaw System is undoubtedly accepted private international air law treaty and has played major role on the carrier's liability in international aviation transport industry. But the whole Warsaw System, though it was revised many times to meet the rapid developments of the aviation transport industry, is so complicated, tangled and outdated. This thesis, therefore, aim to introduce the Montreal Convention by interpreting it as a new legal instrument on the air carrier's liability, specially on the passenger's, and analyzing all the issues relating to it. The Montreal Convention markedly changed the rules governing international carriage by air. The Montreal Convention has modernized and consolidated the old Warsaw System of international instruments of private international air law into one legal instrument. One of the most significant features of the Montreal Convention is that it sifted its priority to the protection of the interest of the consumers from the protection of the carrier which originally the Warsaw Convention intended to protect the fledgling international air transport business. Two major features of the Montreal Convention adopts are the Two-tier Liability System and the Fifth Jurisdiction. In case of death or bodily injury to passengers, the Montreal Convention introduces a two-tier liability system. The first tier includes strict liability up to 100,000SDR, irrespective of carriers' fault. The second tier is based on presumption of fault of carrier and has no limit of liability. Regarding Jurisdiction, the Montreal Convention expands upon the four jurisdiction in which the carrier could be sued by adding a fifth jurisdiction, i.e., a passenger can bring suit in a country in which he or she has their permanent and principal residence and in which the carrier provides a services for the carriage of passengers by either its own aircraft or through a commercial agreement. Other features are introducing the advance payment, electronic ticketing, compulsory insurance and regulation on the contracting and actual carrier etc. As we see some major features of the Montreal Convention, the Convention heralds the single biggest change in the international aviation liability and there can be no doubt it will prevail the international aviation transport world in the future. Our government signed this Convention on 20th Sep. 2007 and it came into effect on 29th Dec. 2007 domestically. Thus, it was recognized that domestic carriers can adequately and independently manage the change of risks of liability. I, therefore, would like to suggest our country's aviation industry including newly-born low cost carrier prepare some countermeasures domestically that are necessary to the enforcement of the Convention.

• Korean Journal of Poultry Science
• /
• v.18 no.2
• /
• pp.97-119
• /
• 1991

Effect of Streptococcus faecium(SF) and an antibiotic, Colistin(Col), supplemented to diets singly or in combination, on the performances and changes of intestinal population of microflora of broiler chicks studied. A total of 252, day-old chicks(Arbor Acre) of mixed sex(M：F＝1：1) were alloted into six groups. A diet with no Col and SF was referred as a control diet. The basal diets were added with two levels of SF, 0.04 and 0.08%, singly or in combination with Col 10ppm Another diet was prepared by adding only Col 10 ppm. Numbers of the microorganism in diets added with SF 0.04% and 0.08% were 7$\times$10$^{4}$ and 1.4$\times$10$^{5}$ /g diet respectively The diets consisting of corn and soybean meal as major ingredients were fed for a period of seven weeks . During the feeding trial, fresh excreta were sampled at the end of every week in a sterilized condition to count microbial changes from each dietary group. Microbial changes of large intestine were also measured from nine birds sacrificed at the end of the 4th and 7th weeks each time per dietary group. Excreta from all the groups were also collected quantitatively at the end of 3rd and 6th weeks to measure digestibility of the diets, At the end of 7th week, nine birds from each group were also sacrificed to measure weight changes of gastrointestinal tracts . Average body weight gains of broilers fed the diets added with SF 0.08% (2.37kg) or SF 0. 08%+col 10ppm(2.34kg) were significantly larger than that of the control(2.18kg). The weight gains of the other groups were not statistically different from that of the control Feed/gain ratios of the supplemental groups were better than that of control (P＜0.05) except that of birds fed the diet added only with SF 0.04%. Digestibilities of nutrients such as dry matter, crude protein, crude fat and total carbohydrates were not altered by the consumption of the diets added with SF and/or Col throughout the whole feeding period. As expected, the numbers of Streptococci in the excreta from birds fed diets added with SF increased significantly with a statistical difference between groups with SF 0.04% and SF 0.08% most of the time. However. addition of Colistin to the diets supplemented with SF did not give any effects on the number of the microorganism. Numbers of coliforms in the excreta were apparently reduced by feeding the diets added with SF and/or Col(P＜0.05). There were, however, no additive effects observed between the two feed additives in this regard when supplementing Col to the SF diets. Distributions of intestinal microflora exhibited exactly the same pattern as those of the excreta. Length of small intestine of the birds fed diets added with SF 0.08% with or without Col 10 ppm became significantly longer with a range of about 10% than those of the birds fed diets without SF. However, the empty weight of the small inestine of the former group was lighter than that of control These changes resulted in a significant reduction in weight/unit length of the intestine of the birds fed diets supplemented with Col and SF singly or in combination. In overall conclusion, diet added with SF 0.08% appeared most effective in improving broiler performances. Colistin added at a level of 10ppm was not beneficial at all in itself or in combination with SF in terms of broiler performances or changes of intestinal microflora population. The efficacy of SF and Col could be attributed to the changes of wall thickness of the small intestine.