• Informatization Policy
• /
• v.24 no.1
• /
• pp.79-99
• /
• 2017

Despite the early success story, the pan-industry diffusion of big data has been slow mostly due to lack of confidence of the value creation and privacy-related concerns. The problem leads us to the need to a stakeholder analysis on the adoption process of big data. The present study combines technology acceptance model, task-technology fit theory, and privacy calculus theory to integrate the positive and negative factors on the big data adoption. The empirical analysis was performed based on the survey from the current and potential big data users. Results revealed perceived usefulness, task-technology fit, and privacy concern are significant antecedents to the intention to use big data. Furthermore, there are significant differences in the perceptions of each constructs among groups divided by the types of big data use, with several exceptions. And the control effect was found in the magnitude of the relation between independent variables and dependent variable. The theoretical and politic implications of the analysis are discussed as to the promotion of big data industry.

• Korean Journal of Remote Sensing
• /
• v.37 no.5_1
• /
• pp.939-946
• /
• 2021

Monitoring of post wildfire provides important information for vegetation restoration. In particular, remote sensing data are known to provide useful information necessary for monitoring. However, there are insufficient research results which is monitoring the vegetation recovery using remote sensing data. This study is directed to monitoring post-wildfire vegetation restoration. It proposes a method for monitoring vegetation restoration using Sentinel-2 satellite data by compositing Tasseled Cap linear regression trend in a post wildfire study sites. Although it is a simple visualization technique using satellite images, it was able to confirm the possibility of effective monitoring.

• Membrane Journal
• /
• v.31 no.4
• /
• pp.268-274
• /
• 2021

One of the chronic problems in the issue of global warming is the emission of greenhouse gases. Carbon dioxide (CO₂), which accounts for the highest proportion of various greenhouse gases, has been continuously researched by humans to separate it. From this point of view, a poly(vinyl alcohol) (PVA)-based copolymer with acrylic acid monomer was utilized in a gas separation membrane in this study. We employed a free radical polymerization to fabricate PVA-g-PAA (VAA) graft copolymer. It was utilized in the form of a composite membrane on a polysulfone substrate. The proper amount of acrylic acid reduced the crystallinity of PVA and increased CO₂ solubility in separation membranes. In this perspective, we suggest the novel approach in CO₂ separation membrane area by grafting and solution-diffusion.

• Information Systems Review
• /
• v.22 no.1
• /
• pp.113-124
• /
• 2020

In the operation of farms, it is very important to evaluate the quality of harvested crops and to classify defective products. However, farmers have difficulty coping with the cost and time required for quality assessment due to insufficient capital and manpower. This study thus aims to detect defects by analyzing the epidermis of fruit using deep learning algorithm. We developed a model that can analyze the epidermis by applying YOLOv3 algorithm based on Region Convolutional Neural Network to video images of peach. A total of four classes were selected and trained. Through 97,600 epochs, a high performance detection model was obtained. The crop failure detection model proposed in this study can be used to automate the process of data collection, quality evaluation through analyzed data, and defect detection. In particular, we have developed an analytical model for peach, which is the most vulnerable to external wounds among crops, so it is expected to be applicable to other crops in farming.

• 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.

• The Journal of the Korea Contents Association
• /
• v.18 no.5
• /
• pp.311-318
• /
• 2018

This paper presents the artistic background and implementation structure of a conversational artificial intelligence interactive artwork, "An Evolving GAIA". Recent artworks based on artificial intelligence technology are introduced. Development of biomimetics and artificial life technology has burred differentiation of machine and human. In this paper, artworks presenting machine-life metaphor are shown, and the distinct implementation of conversation system is emphasized in detail. The artwork recognizes and follows the movement of audience using its eyes for natural interaction. It listens questions of the audience and replies appropriate answers by text-to-speech voice, using the conversation system implemented with an Android client in the artwork and a webserver based on the question-answering dictionary. The interaction gives to the audience discussion of meaning of life in large scale and draws sympathy for the artwork itself. The paper shows the mechanical structure, the implementation of conversational system of the artwork, and reaction of the audience which can be helpful to direct and make future artificial intelligence interactive artworks.

• Journal of Game and Entertainment
• /
• v.2 no.3
• /
• pp.24-29
• /
• 2006

Nowadays, 'Interaction' is discussed not only in the multimedia field but also as a topic of daily life at the various viewpoints. And as the permmited limit of Interaction by Mutimedia Technology and audiovisual function is getting more activated, It is the time for us to have new understanding of a 'POSTER'. The Poster, as a one-sided information transmission media till now, is faced with its limitation in transmitting messages due to its unconformity between satisfaction of receiver' sdesire for information and installation space. This fact speaks for need of study about Digital Poster as an alternative communication method, and activation of Digital Poster Design Field. Therefore, I would like to present Design process and Examples by using Digital Poster for Maximizing the Communication Effect. And I expect the activation of Digital Poster Field by studying the possibility of development and various method of practical use of digital posters through presenting the examples.

• Korean Journal of Logic
• /
• v.22 no.2
• /
• pp.309-332
• /
• 2019

In Kang(2017), Jinho Kang persuasively criticized the attempt of Peter Hanks using his concept of cancelled predication to solve the Frege-Geach problem. According to Kang, Hanks had successfully shown the incoherence of Scott Soames's concept of neutral predication, but if it is true, then Hanks's concept of cancelled predication also falls into the same incoherence. I agree with Kang that Hanks faces the same incoherence, and I think that Hank's answers are unconvincing. As I see, however, it is possible for Hanks to overcome Kang's criticism. In this paper, I will reply to Kang's criticism by using conceptual resources in Hanks's own theory. In particular, the idea is that the final predication is compositionally explained by the type it belongs to without having truth-values because cancelled predication itself gives rise to target-shifting toward the type. By doing so, Hanks can successfully solve the Frege-Geach problem even though he let some remarks about cancelled predication unclear and confusing. In addition, it will be revealed that his notation is misleading as well.

• Journal of the Korea Convergence Society
• /
• v.12 no.2
• /
• pp.67-75
• /
• 2021

The purpose of this study was to investigate the effect of sensory-motor centered group occupational therapy program on self-regulation and executive function in first grade elementary school maladjusted children. It is a case study through pre-post design with a total of 3 subjects. A pre-and post-test was conducted to determine the change in Self-Control Rating Scale and to find out the execution function by Bruininks-Oseretasky Test of Motor Proficiency. The sensory-motor centered group occupational therapy program performed movement activities based on vestibular sensation, proprioception, and tactile sensation, and the task was selected by investigating the child's preference for activity. As a result, subjects 1 and 2 children showed improved self-regulation and executive function. Based on the results of this study, it is considered that the group-centered sensory-motor program can be provided to children who show maladjustment in school by linking the educational field and clinical practice.

• Journal of the Korean earth science society
• /
• v.30 no.2
• /
• pp.196-209
• /
• 2009

Occurrences of sandstorms in the deserts and loess of Mongolia and northern China and associated dustfall episodes in the Korean Peninsula were monitored during the period January through December, 2005. False colour images were made by directly receiving the NOAA Advanced Very High Resolution Radiometer (AVHRR) data, and the distribution and transport of sandstorms were analyzed. The ground concentrations for PM10, PM2.5 and visibility of the dustfall episodes (PM10 concentration over $190{\mu}g\;m^{-3}$) were analyzed at Cheongwon, located midway in South Korea, and in the leeward direction of the place of origin of the sandstorms. Variations in the concentrations of $O_3,\;NO_2$, CO and $SO_2$ were also compared with dust concentrations in the dustfall episodes. Fewer occurrences of strong sandstorms in the place of origin were observed in 2005, due largely to the accumulation of snow and mild fluctuations of high and low pressure systems in the place of origin, thereby accounting for a low frequency of dustfall episodes in Korea, compared with those during the period 1997-2005. A total of 7 dustfall episodes were monitored in Korea in 2005 that lasted 11 days. In summer, sandstorms occurred less frequently in the source region in 2005 due to high humidity and milder winds, thereby causing no dustfall episodes in Korea. In case the sandstorms occurring at the place of source head directly to Korea without passing through large cities and industrial areas of China, the PM2.5 concentrations were measured at 20% or lower than the PM10 concentrations. However, when the sandstorms headed to Korea via the industrial areas of eastern China, where they pick up anthropogenic air pollutants, the PM2.5 concentrations were at least 25% higher of the PM10 concentrations. On the other hand, over 5 cases were observed and analyzed in 2005 where the PM10 concentrations of sand dust originating from the deserts were measured at $190{\mu}g\;m^{-3}$ or lower, falling short of the level of a dustfall episode.