• Journal of Intelligence and Information Systems
• /
• v.26 no.4
• /
• pp.27-65
• /
• 2020

Many companies on information and communication technology make public their own developed AI technology, for example, Google's TensorFlow, Facebook's PyTorch, Microsoft's CNTK. By releasing deep learning open source software to the public, the relationship with the developer community and the artificial intelligence (AI) ecosystem can be strengthened, and users can perform experiment, implementation and improvement of it. Accordingly, the field of machine learning is growing rapidly, and developers are using and reproducing various learning algorithms in each field. Although various analysis of open source software has been made, there is a lack of studies to help develop or use deep learning open source software in the industry. This study thus attempts to derive a strategy for adopting the framework through case studies of a deep learning open source framework. Based on the technology-organization-environment (TOE) framework and literature review related to the adoption of open source software, we employed the case study framework that includes technological factors as perceived relative advantage, perceived compatibility, perceived complexity, and perceived trialability, organizational factors as management support and knowledge & expertise, and environmental factors as availability of technology skills and services, and platform long term viability. We conducted a case study analysis of three companies' adoption cases (two cases of success and one case of failure) and revealed that seven out of eight TOE factors and several factors regarding company, team and resource are significant for the adoption of deep learning open source framework. By organizing the case study analysis results, we provided five important success factors for adopting deep learning framework: the knowledge and expertise of developers in the team, hardware (GPU) environment, data enterprise cooperation system, deep learning framework platform, deep learning framework work tool service. In order for an organization to successfully adopt a deep learning open source framework, at the stage of using the framework, first, the hardware (GPU) environment for AI R&D group must support the knowledge and expertise of the developers in the team. Second, it is necessary to support the use of deep learning frameworks by research developers through collecting and managing data inside and outside the company with a data enterprise cooperation system. Third, deep learning research expertise must be supplemented through cooperation with researchers from academic institutions such as universities and research institutes. Satisfying three procedures in the stage of using the deep learning framework, companies will increase the number of deep learning research developers, the ability to use the deep learning framework, and the support of GPU resource. In the proliferation stage of the deep learning framework, fourth, a company makes the deep learning framework platform that improves the research efficiency and effectiveness of the developers, for example, the optimization of the hardware (GPU) environment automatically. Fifth, the deep learning framework tool service team complements the developers' expertise through sharing the information of the external deep learning open source framework community to the in-house community and activating developer retraining and seminars. To implement the identified five success factors, a step-by-step enterprise procedure for adoption of the deep learning framework was proposed: defining the project problem, confirming whether the deep learning methodology is the right method, confirming whether the deep learning framework is the right tool, using the deep learning framework by the enterprise, spreading the framework of the enterprise. The first three steps (i.e. defining the project problem, confirming whether the deep learning methodology is the right method, and confirming whether the deep learning framework is the right tool) are pre-considerations to adopt a deep learning open source framework. After the three pre-considerations steps are clear, next two steps (i.e. using the deep learning framework by the enterprise and spreading the framework of the enterprise) can be processed. In the fourth step, the knowledge and expertise of developers in the team are important in addition to hardware (GPU) environment and data enterprise cooperation system. In final step, five important factors are realized for a successful adoption of the deep learning open source framework. This study provides strategic implications for companies adopting or using deep learning framework according to the needs of each industry and business.

• Journal of forensic and investigative science
• /
• v.2 no.2
• /
• pp.5-20
• /
• 2007

During the period from 2001 to 2005, 29,892 burglaries took place in Korea with the approximate average annual number - 5,978 cases. This study was conducted to analyze the reported burglaries and the result was summarized as follows. There were 8,605 residential invasion burglaries (28.8%) as the most frequent characteristic pattern. The exit was used as the most frequent invasion route for 4,031 invasion burglaries (64.3%), and an unlocked exit door or window was used as the most frequent invasion method for 2,462 invasion burglaries (28.6%). The hours just after midnight (between 00:00 and 04:00) were the most frequent time for invasion burglary to occur. Also, 5,652 burglaries occurred on Wednesday which was twice higher than on Sunday (2,988 burglaries). It was shown that the number of persons injured during burglaries were 260 deaths and 10,610 injuries. The places of the highest occurrence were the street with 10,183 burglaries (34%) and then residential place with 7,527 burglaries (approximately 25%). One-man burglary was the highest complicity: 15,012 offenders (56.1%). The knife was used as the most frequent instrument for 6,498 burglaries (24,3%) what is rare, while no criminal tool or instrument was used for 15,631 burglaries (58.4). During the period from 2001 to 2006, 1,506 burglaries occurred in Daegu and the average annual number was 251 burglaries. Among those,515 residential invasion burglaries (34.2%) took place and the average annual number was approximately 86 cases. The hours just after midnight (between 00:00 and 04:00) were the most frequent time for invasion burglary to occur (194 cases, 37.7%), the place of the highest invasion occurrence was the residential place (259 cases, 50.3%), and the exit was used as the most frequent invasion route (87 cases, 37.7%). An unlocked exit door or window was the most frequent invasion method (65 cases, 25.1%). In addition, pretending to be a delivery man, visitor or following the victim methods were used for 26 burglaries (10%). It is apparent that personal preventive measures against crimes, as well as governmental and social measures, play an important role in preventing burglaries. In particular, based on the analyzed result that an unlocked window or exit door was most frequently used for reported burglaries, it seems that there is a lack of understanding of crime prevention while little effort has been made to prevent crimes. Although everyone knows that locking a door is one of the basic measures to prevent crimes, many people tend to pay little attention to lock a door properly so burglary takes place. This study, therefore, is intended to encourage people to pay more careful attention to crime prevention, in order to help reduce the probability of burglary. With the recent improvement in social understanding of scientific crime investigation, a wide variety of police professions, including crime analysts, crime victim police counselors and coroners, have been prepared to develop the scientific crime investigation and crime analysis. In addition, it is hoped that further this study will contribute to encourage studies on crime prevention to be carried out in the future.

• Asia pacific journal of information systems
• /
• v.23 no.3
• /
• pp.25-53
• /
• 2013

This study analyzes the effect of Computer Aided Innovation (CAI) to improve R&D Capabilities empirically. Survey was distributed by e-mail and Google Docs, targeting CTO of 235 SMEs. 142 surveys were returned back (rate of return 60.4%) from companies. Survey results from 119 companies (83.8%) which are effective samples except no-response, insincere response, estimated value, etc. were used for statistics analysis. Companies with less than 50billion KRW sales of entire researched companies occupy 76.5% in terms of sample traits. Companies with less than 300 employees occupy 83.2%. In terms of the type of company business Partners (called 'partners with big companies' hereunder) who work with big companies for business occupy 68.1%. SMEs based on their own business (called 'independent small companies') appear to occupy 31.9%. The present status of holding IT system according to traits of company business was classified into partners with big companies versus independent SMEs. The present status of ERP is 18.5% to 34.5%. QMS is 11.8% to 9.2%. And PLM (Product Life-cycle Management) is 6.7% to 2.5%. The holding of 3D CAD is 47.1% to 21%. IT system-holding and its application of independent SMEs seemed very vulnerable, compared with partner companies of big companies. This study is comprised of IT infra and IT Utilization as CAI capacity factors which are independent variables. factors of R&D capabilities which are independent variables are organization capability, process capability, HR capability, technology-accumulating capability, and internal/external collaboration capability. The highest average value of variables was 4.24 in organization capability 2. The lowest average value was 3.01 in IT infra which makes users access to data and information in other areas and use them with ease when required during new product development. It seems that the inferior environment of IT infra of general SMEs is reflected in CAI itself. In order to review the validity used to measure variables, Factors have been analyzed. 7 factors which have over 1.0 pure value of their dependent and independent variables were extracted. These factors appear to explain 71.167% in total of total variances. From the result of factor analysis about measurable variables in this study, reliability of each item was checked by Cronbach's Alpha coefficient. All measurable factors at least over 0.611 seemed to acquire reliability. Next, correlation has been done to explain certain phenomenon by correlation analysis between variables. As R&D capabilities factors which are arranged as dependent variables, organization capability, process capability, HR capability, technology-accumulating capability, and internal/external collaboration capability turned out that they acquire significant correlation at 99% reliability level in all variables of IT infra and IT Utilization which are independent variables. In addition, correlation coefficient between each factor is less than 0.8, which proves that the validity of this study judgement has been acquired. The pair with the highest coefficient had 0.628 for IT utilization and technology-accumulating capability. Regression model which can estimate independent variables was used in this study under the hypothesis that there is linear relation between independent variables and dependent variables so as to identify CAI capability's impact factors on R&D. The total explanations of IT infra among CAI capability for independent variables such as organization capability, process capability, human resources capability, technology-accumulating capability, and collaboration capability are 10.3%, 7%, 11.9%, 30.9%, and 10.5% respectively. IT Utilization exposes comprehensively low explanatory capability with 12.4%, 5.9%, 11.1%, 38.9%, and 13.4% for organization capability, process capability, human resources capability, technology-accumulating capability, and collaboration capability respectively. However, both factors of independent variables expose very high explanatory capability relatively for technology-accumulating capability among independent variable. Regression formula which is comprised of independent variables and dependent variables are all significant (P<0.005). The suitability of regression model seems high. When the results of test for dependent variables and independent variables are estimated, the hypothesis of 10 different factors appeared all significant in regression analysis model coefficient (P<0.01) which is estimated to affect in the hypothesis. As a result of liner regression analysis between two independent variables drawn by influence factor analysis for R&D capability and R&D capability. IT infra and IT Utilization which are CAI capability factors has positive correlation to organization capability, process capability, human resources capability, technology-accumulating capability, and collaboration capability with inside and outside which are dependent variables, R&D capability factors. It was identified as a significant factor which affects R&D capability. However, considering adjustable variables, a big gap is found, compared to entire company. First of all, in case of partner companies with big companies, in IT infra as CAI capability, organization capability, process capability, human resources capability, and technology capability out of R&D capacities seems to have positive correlation. However, collaboration capability appeared insignificance. IT utilization which is a CAI capability factor seemed to have positive relation to organization capability, process capability, human resources capability, and internal/external collaboration capability just as those of entire companies. Next, by analyzing independent types of SMEs as an adjustable variable, very different results were found from those of entire companies or partner companies with big companies. First of all, all factors in IT infra except technology-accumulating capability were rejected. IT utilization was rejected except technology-accumulating capability and collaboration capability. Comprehending the above adjustable variables, the following results were drawn in this study. First, in case of big companies or partner companies with big companies, IT infra and IT utilization affect improving R&D Capabilities positively. It was because most of big companies encourage innovation by using IT utilization and IT infra building over certain level to their partner companies. Second, in all companies, IT infra and IT utilization as CAI capability affect improving technology-accumulating capability positively at least as R&D capability factor. The most of factor explanation is low at around 10%. However, technology-accumulating capability is rather high around 25.6% to 38.4%. It was found that CAI capability contributes to technology-accumulating capability highly. Companies shouldn't consider IT infra and IT utilization as a simple product developing tool in R&D section. However, they have to consider to use them as a management innovating strategy tool which proceeds entire-company management innovation centered in new product development. Not only the improvement of technology-accumulating capability in department of R&D. Centered in new product development, it has to be used as original management innovative strategy which proceeds entire company management innovation. It suggests that it can be a method to improve technology-accumulating capability in R&D section and Dynamic capability to acquire sustainable competitive advantage.

• Korean Journal of Food Science and Technology
• /
• v.47 no.1
• /
• pp.56-62
• /
• 2015

Various pretreatment methods were evaluated to prevent tissue softening of heated onion. Changes in onion tissue firmness during heating were explained by 3-mechanism model consisting of texture hardening at low temperature ($60-80^{\circ}C$) and substrate softening at high temperature. Preheating of onion in a $Ca^{2+}$-containing solution significantly improved its texture after high-temperature heating. The improvement of firmness by preheating at low temperature was related to the formation of strong cross-linking between carboxyl groups and $Ca^{2+}$ by the action of pectin methylesterase in onion. The highest firmness was obtained by pre-heating at $70^{\circ}C$ for 120 min in 0.5% calcium solution. This result was supported by chemical analysis showing that the amount of bound calcium was the highest at $70^{\circ}C$. Further investigation should be carried out to establish the optimal conditions to prevent the softening of various vegetables.

• The Korean Journal of Nuclear Medicine Technology
• /
• v.18 no.2
• /
• pp.39-47
• /
• 2014

Purpose As breast cancer patients continue to increase every year, cases of BSGI are on the rise with a heavier reliance on it. However, BSGI protocol in hospitals was not studied enough despite it was covered by hospital's condition and recommendation of manufacturers. The objective of the study was an examination of methods to be applicable to BSGI protocols, putting the self-development phantom to use in quality assessment of the images. Materials and Methods Dilon 6800 (Dilon Technologies Inc, Newport News, USA) was used in the study and five different sizes of sphere were distinctively produced in the phantom. The study used $^{99m}TcO_4$. The cases were classified in to three categories that background radioactivity to region of interest as ratio of 2: 4: 8, They were acquired images for 5, 7, 10mins. The acquired image was set region of interest according to the size of sphere, and We analyzed quantitative and qualitative analysis. The acquired data statistically analyzed with SPSS ver.18.0. Results As the result of quantitative and qualitative analysis, count rate of each sphere in accordance with difference of injection dose showed that higher count rate as injection dose and sphere size increased (P<0.005). Count rate of each sphere in accordance with difference of acquisition time showed that higher count rate as acquisition time and sphere size increased (P<0.005). Contrast noise ratio of each sphere in accordance with difference of injection dose showed that higher contrast noise ratio as injection dose increased. Particularly, Contrast noise ratio of eight times ratio images was the highest among. Contrast noise ratio of each sphere in accordance with difference of acquisition time showed that higher contrast noise ratio as acquisition time increased. And, Contrast noise ratio of seven minute image was the highest among (P<0.005). Conclusion There was significant change of Contrast noise ratio through quantitative and qualitative analysis. Moreover, We found usefulness of phantom. If Institutions identified image through the phantom study and they made BSGI protocol, We expected to help the improvement of diagnostic value of the images.

• Journal of Nutrition and Health
• /
• v.42 no.1
• /
• pp.5-13
• /
• 2009

Mushrooms have become a largely untapped source of powerful new pharmaceutical products that poses anti-inflammatory, and antimutagenic, and antioxidant activities. The antioxidant effects of the mushroom may be partly explained by protecting cellular components against free radical. The aim of this study was to investigate the protective effect of chaga mushroom against diabetes, via the mitigation of oxidative stress and reduction of blood glucose, in streptozotocin-induced diabetic rats. Rats were rendered diabetic by intravenous administration of STZ through tail at a dose of 50 mg/kg. Animals were allocated into four groups with 8 rats each. The control and diabetic control group were fed with standard rat feed. The other diabeic groups, the low chaga extract group and the high chaga extract group were fed ad libitum using 0.5 g/kg and 5 g/kg of chaga mushroom extract, respectively, for 4 weeks. The blood glucose levels in the two chaga extract groups showed a tendency to decrease but did not reach statistical significance after the supplementation. Leukocyte DNA damage, expressed as tail length, was found to be significantly lower in the high chaga extract group than in the diabetic control group (p > 0.05). Plasma level of total radical-trapping antioxidant potential (TRAP) was tend to be higher in the high chaga extract group compared with the diabetic control group. Erythrocyte antioxidant enzyme activities of two groups did not differ. Although we did not obtain beneficial effect on lowering blood glucose levels in the STZ-induced diabetic rats, this results suggest that the chaga mushroom extracts may initially act on protecting endogenous DNA damage in the short-term experiment.

• Journal of Nutrition and Health
• /
• v.53 no.4
• /
• pp.416-430
• /
• 2020

Purpose: This study was conducted to assess the effects of a personalized nutritional intervention program on food security and health and nutritional status of elderly people in the city of Seoul. Methods: A total of 372 elderly adults aged 65 years or above who resided in Seoul were enrolled in this study. Personalized supplementary food supply and nutritional education based on chronic disease status, disability and cooking ability were implemented for 4 months. To evaluate the effectiveness of the program, nutrient intake, food security status, anemia status, chronic disease management, and frailty status, and prevalence of malnutrition (Mini Nutritional Assessment) were examined. Results: After the program, all subjects displayed significantly increased nutrient intake. Before the intervention, all subjects were in a state of food insecurity; however, after the intervention, 37.1% of the subjects were food secure. Moreover, the rates of being at risk of malnutrition and malnutrition in subjects were decreased and instead rate of those who improved to normal increased to 29.8% from 0% of normal rate before the prevention. The rate of subjects without anemia increased from 18.7% to 28.5% after the intervention. In addition, the rate of subjects with intensive or periodic management of chronic diseases decreased, while those with occasional management of chronic diseases increased from 0% to 4.6%. Furthermore, the rates of being at risk of frailty and frailty were decreased and the normal rate increased from 0% to 9.7% instead. Age group-based analysis showed that elderly people over 80 years showed less improvement in the management of the chronic disease status and the frailty status. Conclusion: Personalized supplementary food supplies and nutritional education improved not only the nutritional status but also disease status in vulnerable older adults, and the effects were more significant in adults aged less than 80 years.

• Asia-Pacific Journal of Business Venturing and Entrepreneurship
• /
• v.7 no.4
• /
• pp.101-112
• /
• 2012

Business consulting which aims at advising overall management, comparative advantage in competitive environment, administration and so forth is deemed to be significant so that a credit union may flexibly respond to rapid and complicated financial environment. Accordingly, business consulting has been initiated and implemented in credit union since 2002, which seemingly leads to positive outcome to enhance its competitiveness. However, very few research exists which analyze the credit union's competitiveness and business performance after the business consulting. Thus, the research intends to verify how consulting influences on management performance by its various factors. With the above studies in mind, hypothesis is formed that the characteristics of a consulting comsumer and the characteristics of supplier will give a meaningful impact on management performance. The test shows the result as followings: Firstly, a primary credit union's trust and NACUFOK's expertise demonstrate that they are closely relevant to financial and non financial management performance. Secondly, support and willingness to apply states that they influences on non financial performance, while no direct impact is found to financial performance. Thirdly, NACUFOK's capability is rarely interrelated to financial and non financial performance. The present study implies that the credit union as a consulting beneficiary should participate in the consulting, accept the result, and make effort to solve the management problems. In addition, NACUFOK as a consulting supplier should initiate the organization in exclusive charge, perform the professional group, and foster the professional consultant for the development of consulting method, the improvement of consulting process, and the specialization of consultant including the enhancement of core competencies. When these perspectives are sufficient, it will be connected to the management outcomes, further, it will be boosted up the competitiveness of the Credit Union. Hopefully the present study helps to motivate the consulting of primary credit union, boost up the effectiveness of consulting, further, providing the perspective on consulting the credit union. And these may result in the indication of consulting development direction and the competitiveness of Credit Union.

• Journal of the korean academy of Pediatric Dentistry
• /
• v.39 no.4
• /
• pp.383-388
• /
• 2012

The aim of this study was to compare the compressive strength and the surface microhardness of Beautifil flow (Shofu, Kyoto, Japan) with $Filtek^{TM}$ Z350, Z350XT (3M ESPE, USA). Fifteen specimens from each material were fabricated for testing. Compressive strength was measured by using a universal testing machine at a crosshead speed of 1 mm/min. Surface microhardness values were measured by using Vickers hardness tester under 4.9 N load and 10 seconds dwelling time. The compressive strength of Group 2 $Filtek^{TM}$ Z350XT shows the highest value as $218.7{\pm}18.4$ MPa and Group 1 $Filtek^{TM}$ Z350 was $205.5{\pm}27.1$ MPa. Group 3 Beautifil flow F00 was $176.5{\pm}30.3$ MPa, and Group 4 Beautifil flow F10 was $173.4{\pm}26.2$ MPa. The compressive strength of Group 2 is higher than Group 3 and 4 (p < 0.05). The surface microhardness of Group 2 $Filtek^{TM}$ Z350XT shows the highest value as $39.1{\pm}2.1$ and Group 4 Beautifil flow F10 was $27.9{\pm}1.8$. And Group 3 Beautifil flow F00 was $23.1{\pm}1.1$, Group 1 $Filtek^{TM}$ Z350 was $20.4{\pm}0.9$. There was a statistical significant difference in surface microhardness between all groups (p < 0.05). In conclusion, the compressive strength of giomer was below the level of flowable composite resin. However, the surface microhardness of giomer is comparable to that of flowable composite resin. Giomer would be the good alternative to composite resin, if there is improvement of the compressive strength of giomer.

• Journal of Life Science
• /
• v.19 no.8
• /
• pp.1145-1151
• /
• 2009

This study was performed to investigate the effects of Sineui-Dae (Sasa coreana Nakai, a kind of bamboo) extracts on the improvement of serum lipid composition by using rats fed a hypercholesterol diet for 4 weeks. The experiment animals were administered with the following diets; high cholesterol diet group (HC diet) as a control and three supplemented groups with high cholesterol diets (HCW, HCM or HCH diet). Three kinds of extracts were prepared by orderly extraction with hexane, methanol, and water. We measured free cholesterol, cholesteryl ester, total cholesterol LDL- and HDL-cholesterol triglyceride and phospholipid in the serum of rats in three experimental and control groups. There were no significant differences in body weights and feed intakes between the HC control group and the extract supplemented groups. The levels of total cholesterol and LDL-cholesterol in the serum of extract supplemented groups were lower than that of the group fed with HC diet only. Serum HDL-cholesterol levet which is known as an antiatherosclerosis factor, was higher in all groups supplemented with bamboo extract by 17.2-21.9% compared to the HC control group. All groups supplemented with bamboo extracts showed the lowering effect of atherogenic index compared to the HC control group (HC group: 2.96${\pm}$0.08, HCM group: 1.48${\pm}$0.02, HCW group: 1.69${\pm}$0.04, HCH group: 1.84${\pm}$0.01). Furthermore, serum triglyceride and phospholipid decreased significantly in the HCM diet compared to the HC control diet. These results suggest that Sineui-Dae bamboo extract, especially methanol extract, has improving effects on hyperlipidemia of rats fed a high cholesterol diet.