• Journal of the Korean Institute of Intelligent Systems
• /
• v.26 no.2
• /
• pp.93-98
• /
• 2016

For accurate precipitation forecasts the choice of weather factors and prediction method is very important. Recently, machine learning has been widely used for forecasting precipitation, and artificial neural network, one of machine learning techniques, showed good performance. In this paper, we suggest a new method for forecasting precipitation using DBN, one of deep learning techniques. DBN has an advantage that initial weights are set by unsupervised learning, so this compensates for the defects of artificial neural networks. We used past precipitation, temperature, and the parameters of the sun and moon's motion as features for forecasting precipitation. The dataset consists of observation data which had been measured for 40 years from AWS in Seoul. Experiments were based on 8-fold cross validation. As a result of estimation, we got probabilities of test dataset, so threshold was used for the decision of precipitation. CSI and Bias were used for indicating the precision of precipitation. Our experimental results showed that DBN performed better than MLP.

• Asia-Pacific Journal of Business Venturing and Entrepreneurship
• /
• v.14 no.5
• /
• pp.79-92
• /
• 2019

When asked about the success factors of successful entrepreneurs and celebrities, he says he was lucky. The remarkable fact is that the attitude about luck is different. However, despite the fact that the belief that we believe is lucky is actually a dominant concept, there has not been much scientific verification of luck. In this study, we saw good luck not being determined randomly by the external environment, but by being able to control luck through the internal attributes of individuals. This study is significant that we have empirically elucidated what kind of efforts have gained good luck, whereas previous research has largely ended in vague logic where luck ends up with an internal locus of control among internal entrepreneurial qualities and efforts can make a successful entrepreneur. We introduced the concept of good luck belief to avoid confirmation bias, which is, to interpret my experience in a direction that matches what I want to believe, and used a good luck belief questionnaire in previous studies and tried to verify that those who have a good belief can increase entrepreneurial opportunity capability through planned happenstance skills. The reason for choosing the entrepreneurial opportunity capacity as a dependent variable was based on the conventional research, that is, the process of recognizing and exploiting the entrepreneurial opportunity is an important part of the entrepreneurship research For empirical research, we conducted a questionnaire survey of a total of 332 people, and the results of the analysis turned out that the belief of good luck has all the positive impacts of planned happenstance skills' sub-factors: curiosity, patience, flexibility, optimism and risk tolerance. Second, we have shown that only the perseverance, optimism, and risk tolerance of planned happenstance skills' sub-factors have a positive impact on this opportunity capability. Thirdly, it was possible to judge that the sub-factors of planned happenstance skills, patience, optimism, and risk tolerance, had a meditating effect between belief in luck and entrepreneurial opportunity capability. This study is highly significant in logically elucidating that people in charge of business incubation and education can get the specific direction when planning a training program for successful entrepreneur to further enhance the entrepreneurial opportunity ability, which is an important ability for the entrepreneur's success.

• Korean Journal of Adult Nursing
• /
• v.28 no.1
• /
• pp.13-29
• /
• 2016

Purpose: This study was performed to evaluate the effects of non-pharmacological interventions on sleep disturbance amongst adults aged 55 and above. Methods: PubMed, Cochrane Library, EMBASE, CINAHL and several Korean databases were searched. The main search strategy combined terms including non-pharmacological interventions and presence of insomnia. Non-pharmacological interventions included cognitive behavioral therapy, auricular acupuncture, aromatherapy, and emotional freedom techniques. Methodological quality was assessed using Cochrane's Risk of Bias for randomized studies and Risk of Bias Assessment tool for non randomized studies. Data were analyzed by the RevMan 5.3 program of Cochrane Library. Results: Sixteen clinical trials met the inclusion criteria with a total of 962 participants. Non-pharmacological interventions was conducted for a mean of 5.5 weeks, 7.7 sessions, and an average of 70 minutes per session. The effects of non-pharmacological interventions on sleep quality (ES=-1.18), sleep efficiency (ES=-1.14), sleep onset latency (ES=-0.88), awakening time after sleep onset (ES=-0.87), and sleep belief (ES=-0.71) were significant, and their effect sizes were ranged from moderate to large. However, the effects on total sleep time and insomnia severity were not significant. Conclusion: The findings of the current study suggest that non-pharmacological interventions have a positive impact on attitudes and beliefs about sleep, sleep quality, sleep duration, and sleep efficiency. Therefore, the findings of the study provide an evidence to incorporate various non-pharmacological interventions into nursing practice to improve both sleep quality and quantity in patients with insomnia.

• Knowledge Management Research
• /
• v.17 no.4
• /
• pp.1-25
• /
• 2016

The main purpose of this study is to examine relations among authentic leadership, trust for leaders, psychological well-being, and knowledge sharing. Authentic leadership proposes positive and interactional approach between leaders and subordinates. Authentic leaders are aware of their values and belief, and they keep their personal goals and support their followers. Such behaviors boost psychological well-being, knowledge sharing, and trust for leaders. To analyze the framework proposed, survey data was collected from 164 employees of three companies. In particular, this study designed a robust research method by assessing model fit and avoiding common method bias issues. The empirical results of this study are as follows. Authentic leadership positively influences trust for leaders and psychological well-being. Trust for leaders is shown to have positive impacts on psychological well-being and knowledge sharing. In addition, followers' psychological well-being positively influences knowledge sharing activities. This study contributes to the comprehension of the structural relationships among authentic leadership, trust for leaders, psychological well-being, and knowledge sharing. The results suggest that authentic leadership and trust for leaders were key success factors of building positive mindset and capability of employees in the forms of psychological well-being and knowledge sharing activities.

• Asian Journal for Public Opinion Research
• /
• v.9 no.1
• /
• pp.3-24
• /
• 2021

Through an empirical comparative analysis, we found that people in mainland China and Taiwan demonstrate strong similarities in their support for democracy, based on democratic suitability, efficiency, preference, and priority. There are also differences in beliefs about democratic values. Compared to people in mainland China, the Taiwanese have a deeper and more widely shared belief in the principles of participation and pluralism, while the differences between their beliefs in the principles of equality, freedom, and checks and balances are narrow. Furthermore, people in mainland China and Taiwan have a strong similarity in their understanding of democracy, that is, they all present a mixed democratic understanding based on substantive bias. Overall, although the differences between mainland China and Taiwan's democratic practices are reflected in the level of value identification from the perspective of democratic support and democratic understanding, the popular democratic political culture in mainland China and Taiwan still has a relatively broad consensus. Thus, the integration and development of cross-strait relations not only has an increasingly profound social and economic foundation but also considerable consensus and mass support on the political and cultural level.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.17 no.9
• /
• pp.407-421
• /
• 2016

This paper reports a meta-analysis of sixteen studies that evaluated the efficacy of cognitive behavioral therapy (CBT) for persistent primary insomnia. PubMed, Cochrane Library, EMBASE, CINAHL and several Korean databases were searched between January 2015 and June 2015. The main search strategy involved the terms that indicate CBT-I (Cognitive Behavioral Therapy-Insomnia) and presence of insomnia. Methodological quality was assessed using Cochrane's Risk of Bias. Data were analyzed by the RevMan 5.3 program of Cochrane Library. Sixteen clinical trials met the inclusion criteria, resulting in a total of 1503 participants. Stimulus control, sleep restriction, sleep hygiene education, and cognitive restructuring were the main treatment components. CBT-I was conducted for a mean of 5.4 weeks, 5.5 sessions, and an average of 90 minutes per session. The effects of CBT-i on total sleep time (d=-0.31), sleep onset latency (d=-0.29), awakening time after sleep onset (d=-0.55), sleep efficiency (d=-0.70), insomnia severity (d=-0.77) and sleep belief (d=-0.64) were significant. Overall, we found a range from small to moderate effect size. CBT-I also was effective for anxiety (d=-0.30) and depression (d=-0.35). The findings demonstrate that CBT-I interventions will lead to the improvement of both sleep quality and quantity in patients with insomnia.

• Structural Engineering and Mechanics
• /
• v.37 no.4
• /
• pp.427-442
• /
• 2011

In fatigue life design of mechanical components, uncertainties arising from materials and manufacturing processes should be taken into account for ensuring reliability. A common practice is to apply a safety factor in conjunction with a physics model for evaluating the lifecycle, which most likely relies on the designer's experience. Due to conservative design, predictions are often in disagreement with field observations, which makes it difficult to schedule maintenance. In this paper, the Bayesian technique, which incorporates the field failure data into prior knowledge, is used to obtain a more dependable prediction of fatigue life. The effects of prior knowledge, noise in data, and bias in measurements on the distribution of fatigue life are discussed in detail. By assuming a distribution type of fatigue life, its parameters are identified first, followed by estimating the distribution of fatigue life, which represents the degree of belief of the fatigue life conditional to the observed data. As more data are provided, the values will be updated to reduce the credible interval. The results can be used in various needs such as a risk analysis, reliability based design optimization, maintenance scheduling, or validation of reliability analysis codes. In order to obtain the posterior distribution, the Markov Chain Monte Carlo technique is employed, which is a modern statistical computational method which effectively draws the samples of the given distribution. Field data of turbine components are exploited to illustrate our approach, which counts as a regular inspection of the number of failed blades in a turbine disk.

• Journal of Korean Library and Information Science Society
• /
• v.24
• /
• pp.485-517
• /
• 1996

The purpose of this study is to investigate typical ethical problems found in the technical and public services areas. The followings are the summary of the study. There are three distinct elements that govern ethical problems. One element is legal laws. The copyright law and the privacy act are exact examples. The copyright law has strong influence on the inter library loan service where the majority requests from the users are reproduction of copies. The privacy act also creates difficulties for librarians. Most requests for circulation records infringe on the privacy of library user. And advance online access systems also violates the privacy of library users. The second element is the code or rules that private organization has created. American Library Association created many statements that regulate the conduct of librarians. The bill of right, the professional code of ethics and policy on the confidentiality of library records have strong implications in the obligation of librarian. In the case of censorship at the selection of library materials, the code is a defensive tool against intellectual freedom. Yet self-censoring are prevailing practice among librarians. The thirds element is the competence of librarians. The analyzed table 3 showed that beside two elements, the rest of matters are competence required by librarians. The one aspect of it is humaneness and the other one is technical aspects. Technical aspect of competence are:(l) managerial and operational ability (2) communication skill (3) leadership (4) structure of knowledge and (5) self developing professionalism. Humanity aspect of competence are:(l) trust(fiduciary relationship) gained by diligence, objective judgement, ability, belief, rationality, integrity, kindness) (2) objectiveness (free from bias) (3) user-oriented consideration (need, interest, equal treatment, information gap) (4) caution in providing information (5) pride and (6) ability to distinguish advice and guidance specially in medical and law library.

• Journal of Intelligence and Information Systems
• /
• v.22 no.2
• /
• pp.127-142
• /
• 2016

Deep learning model is a kind of neural networks that allows multiple hidden layers. There are various deep learning architectures such as convolutional neural networks, deep belief networks and recurrent neural networks. Those have been applied to fields like computer vision, automatic speech recognition, natural language processing, audio recognition and bioinformatics where they have been shown to produce state-of-the-art results on various tasks. Among those architectures, convolutional neural networks and recurrent neural networks are classified as the supervised learning model. And in recent years, those supervised learning models have gained more popularity than unsupervised learning models such as deep belief networks, because supervised learning models have shown fashionable applications in such fields mentioned above. Deep learning models can be trained with backpropagation algorithm. Backpropagation is an abbreviation for "backward propagation of errors" and a common method of training artificial neural networks used in conjunction with an optimization method such as gradient descent. The method calculates the gradient of an error function with respect to all the weights in the network. The gradient is fed to the optimization method which in turn uses it to update the weights, in an attempt to minimize the error function. Convolutional neural networks use a special architecture which is particularly well-adapted to classify images. Using this architecture makes convolutional networks fast to train. This, in turn, helps us train deep, muti-layer networks, which are very good at classifying images. These days, deep convolutional networks are used in most neural networks for image recognition. Convolutional neural networks use three basic ideas: local receptive fields, shared weights, and pooling. By local receptive fields, we mean that each neuron in the first(or any) hidden layer will be connected to a small region of the input(or previous layer's) neurons. Shared weights mean that we're going to use the same weights and bias for each of the local receptive field. This means that all the neurons in the hidden layer detect exactly the same feature, just at different locations in the input image. In addition to the convolutional layers just described, convolutional neural networks also contain pooling layers. Pooling layers are usually used immediately after convolutional layers. What the pooling layers do is to simplify the information in the output from the convolutional layer. Recent convolutional network architectures have 10 to 20 hidden layers and billions of connections between units. Training deep learning networks has taken weeks several years ago, but thanks to progress in GPU and algorithm enhancement, training time has reduced to several hours. Neural networks with time-varying behavior are known as recurrent neural networks or RNNs. A recurrent neural network is a class of artificial neural network where connections between units form a directed cycle. This creates an internal state of the network which allows it to exhibit dynamic temporal behavior. Unlike feedforward neural networks, RNNs can use their internal memory to process arbitrary sequences of inputs. Early RNN models turned out to be very difficult to train, harder even than deep feedforward networks. The reason is the unstable gradient problem such as vanishing gradient and exploding gradient. The gradient can get smaller and smaller as it is propagated back through layers. This makes learning in early layers extremely slow. The problem actually gets worse in RNNs, since gradients aren't just propagated backward through layers, they're propagated backward through time. If the network runs for a long time, that can make the gradient extremely unstable and hard to learn from. It has been possible to incorporate an idea known as long short-term memory units (LSTMs) into RNNs. LSTMs make it much easier to get good results when training RNNs, and many recent papers make use of LSTMs or related ideas.