• Journal of the Institute of Convergence Signal Processing
• /
• v.24 no.1
• /
• pp.42-50
• /
• 2023

The development of technology has achieved digital innovation throughout the media industry, including production techniques and editing technologies, and has brought diversity in the form of consumer viewing through the OTT service and streaming era. The convergence of big data and deep learning networks automatically generated text in format such as news articles, novels, and scripts, but there were insufficient studies that reflected the author's intention and generated story with contextually smooth. In this paper, we describe the flow of pictures in the storyboard with image caption generation techniques, and the automatic generation of story-tailored scenarios through language models. Image caption using CNN and Attention Mechanism, we generate sentences describing pictures on the storyboard, and input the generated sentences into the artificial intelligence natural language processing model KoGPT-2 in order to automatically generate scenarios that meet the planning intention. Through this paper, the author's intention and story customized scenarios are created in large quantities to alleviate the pain of content creation, and artificial intelligence participates in the overall process of digital content production to activate media intelligence.

• Journal of Intelligence and Information Systems
• /
• v.23 no.1
• /
• pp.95-108
• /
• 2017

Recently, AlphaGo which is Bakuk (Go) artificial intelligence program by Google DeepMind, had a huge victory against Lee Sedol. Many people thought that machines would not be able to win a man in Go games because the number of paths to make a one move is more than the number of atoms in the universe unlike chess, but the result was the opposite to what people predicted. After the match, artificial intelligence technology was focused as a core technology of the fourth industrial revolution and attracted attentions from various application domains. Especially, deep learning technique have been attracted as a core artificial intelligence technology used in the AlphaGo algorithm. The deep learning technique is already being applied to many problems. Especially, it shows good performance in image recognition field. In addition, it shows good performance in high dimensional data area such as voice, image and natural language, which was difficult to get good performance using existing machine learning techniques. However, in contrast, it is difficult to find deep leaning researches on traditional business data and structured data analysis. In this study, we tried to find out whether the deep learning techniques have been studied so far can be used not only for the recognition of high dimensional data but also for the binary classification problem of traditional business data analysis such as customer churn analysis, marketing response prediction, and default prediction. And we compare the performance of the deep learning techniques with that of traditional artificial neural network models. The experimental data in the paper is the telemarketing response data of a bank in Portugal. It has input variables such as age, occupation, loan status, and the number of previous telemarketing and has a binary target variable that records whether the customer intends to open an account or not. In this study, to evaluate the possibility of utilization of deep learning algorithms and techniques in binary classification problem, we compared the performance of various models using CNN, LSTM algorithm and dropout, which are widely used algorithms and techniques in deep learning, with that of MLP models which is a traditional artificial neural network model. However, since all the network design alternatives can not be tested due to the nature of the artificial neural network, the experiment was conducted based on restricted settings on the number of hidden layers, the number of neurons in the hidden layer, the number of output data (filters), and the application conditions of the dropout technique. The F1 Score was used to evaluate the performance of models to show how well the models work to classify the interesting class instead of the overall accuracy. The detail methods for applying each deep learning technique in the experiment is as follows. The CNN algorithm is a method that reads adjacent values from a specific value and recognizes the features, but it does not matter how close the distance of each business data field is because each field is usually independent. In this experiment, we set the filter size of the CNN algorithm as the number of fields to learn the whole characteristics of the data at once, and added a hidden layer to make decision based on the additional features. For the model having two LSTM layers, the input direction of the second layer is put in reversed position with first layer in order to reduce the influence from the position of each field. In the case of the dropout technique, we set the neurons to disappear with a probability of 0.5 for each hidden layer. The experimental results show that the predicted model with the highest F1 score was the CNN model using the dropout technique, and the next best model was the MLP model with two hidden layers using the dropout technique. In this study, we were able to get some findings as the experiment had proceeded. First, models using dropout techniques have a slightly more conservative prediction than those without dropout techniques, and it generally shows better performance in classification. Second, CNN models show better classification performance than MLP models. This is interesting because it has shown good performance in binary classification problems which it rarely have been applied to, as well as in the fields where it's effectiveness has been proven. Third, the LSTM algorithm seems to be unsuitable for binary classification problems because the training time is too long compared to the performance improvement. From these results, we can confirm that some of the deep learning algorithms can be applied to solve business binary classification problems.

• 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.

• Journal of Intelligence and Information Systems
• /
• v.24 no.4
• /
• pp.219-240
• /
• 2018

Sentiment analysis, which is one of the text mining techniques, is a method for extracting subjective content embedded in text documents. Recently, the sentiment analysis methods have been widely used in many fields. As good examples, data-driven surveys are based on analyzing the subjectivity of text data posted by users and market researches are conducted by analyzing users' review posts to quantify users' reputation on a target product. The basic method of sentiment analysis is to use sentiment dictionary (or lexicon), a list of sentiment vocabularies with positive, neutral, or negative semantics. In general, the meaning of many sentiment words is likely to be different across domains. For example, a sentiment word, 'sad' indicates negative meaning in many fields but a movie. In order to perform accurate sentiment analysis, we need to build the sentiment dictionary for a given domain. However, such a method of building the sentiment lexicon is time-consuming and various sentiment vocabularies are not included without the use of general-purpose sentiment lexicon. In order to address this problem, several studies have been carried out to construct the sentiment lexicon suitable for a specific domain based on 'OPEN HANGUL' and 'SentiWordNet', which are general-purpose sentiment lexicons. However, OPEN HANGUL is no longer being serviced and SentiWordNet does not work well because of language difference in the process of converting Korean word into English word. There are restrictions on the use of such general-purpose sentiment lexicons as seed data for building the sentiment lexicon for a specific domain. In this article, we construct 'KNU Korean Sentiment Lexicon (KNU-KSL)', a new general-purpose Korean sentiment dictionary that is more advanced than existing general-purpose lexicons. The proposed dictionary, which is a list of domain-independent sentiment words such as 'thank you', 'worthy', and 'impressed', is built to quickly construct the sentiment dictionary for a target domain. Especially, it constructs sentiment vocabularies by analyzing the glosses contained in Standard Korean Language Dictionary (SKLD) by the following procedures: First, we propose a sentiment classification model based on Bidirectional Long Short-Term Memory (Bi-LSTM). Second, the proposed deep learning model automatically classifies each of glosses to either positive or negative meaning. Third, positive words and phrases are extracted from the glosses classified as positive meaning, while negative words and phrases are extracted from the glosses classified as negative meaning. Our experimental results show that the average accuracy of the proposed sentiment classification model is up to 89.45%. In addition, the sentiment dictionary is more extended using various external sources including SentiWordNet, SenticNet, Emotional Verbs, and Sentiment Lexicon 0603. Furthermore, we add sentiment information about frequently used coined words and emoticons that are used mainly on the Web. The KNU-KSL contains a total of 14,843 sentiment vocabularies, each of which is one of 1-grams, 2-grams, phrases, and sentence patterns. Unlike existing sentiment dictionaries, it is composed of words that are not affected by particular domains. The recent trend on sentiment analysis is to use deep learning technique without sentiment dictionaries. The importance of developing sentiment dictionaries is declined gradually. However, one of recent studies shows that the words in the sentiment dictionary can be used as features of deep learning models, resulting in the sentiment analysis performed with higher accuracy (Teng, Z., 2016). This result indicates that the sentiment dictionary is used not only for sentiment analysis but also as features of deep learning models for improving accuracy. The proposed dictionary can be used as a basic data for constructing the sentiment lexicon of a particular domain and as features of deep learning models. It is also useful to automatically and quickly build large training sets for deep learning models.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.19 no.10
• /
• pp.410-417
• /
• 2018

System design errors are more likely to occur in modern systems because of their steadily increasing size and complexity. Failures due to system design errors can cause safety-related accidents in the system, resulting in extensive damage to people and property. Therefore, international standards organizations, such as the U.S. Department of Defense and the International Electrotechnical Commission, have established international safety standards to ensure system safety, and recommend that system design and safety activities should be integrated. Recently, the safety of a system has been verified by modeling through a model-based system design. On the other hand, system design and safety activities have not been integrated because the model for system design and the failure model for safety analysis and verification were developed using different modeling language platforms. Furthermore, studies using UML or SysML-based failure models for deriving safety requirements have shown that these models have limited applicability to safety analysis and verification. To solve this problem, it is essential to extend the existing methods for failure model implementation. First, an improved SysML-based failure model capable of integrating system design and safety verification activities should be produced. Next, this model should help verify whether the safety requirements derived via the failure model are reflected properly in the system design. Therefore, this paper presents the concept and method of developing a SysML-based failure model for an automotive system. In addition, the failure model was simulated to verify the safety of the automotive system. The results show that the improved SysML-based failure model can support the integration of system design and safety verification activities.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.22 no.5
• /
• pp.465-473
• /
• 2002

Images are easily corrupted by noise during the data transmission, data capture and data processing. A technical method of noise analyzing and adaptive filtering for reducing of quantum noise in radiography is presented. By adjusting the characteristics of the filter according to local statistics around each pixel of the image as moving windowing, it is possible to suppress noise sufficiently while preserve edge and other significant information required in reading. We have proposed adaptive weighted median(AWM) filters based on local statistics. We show two ways of realizing the AWM filters. One is a simple type of AWM filter, whose weights are given by a simple non-linear function of three local characteristics. The other is the AWM filter which is constructed by homogeneous factor(HF). Homogeneous factor(HF) from the quantum noise models that enables the filter to recognize the local structures of the image is introduced, and an algorithm for determining the HF fitted to the detection systems with various inner statistical properties is proposed. We show by the experimented that the performances of proposed method is superior to these of other filters and models in preserving small details and suppressing the noise at homogeneous region. The proposed algorithms were implemented by visual C++ language on a IBM-PC Pentium 550 for testing purposes, the effects and results of the noise filtering were proposed by comparing with images of the other existing filtering methods.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.12
• /
• pp.53-59
• /
• 2021

In this paper, we propose a BERGPT-chatbot, a domestic AI chatbot that can alleviate negative emotions based on text input such as 'Replika'. We made BERGPT-chatbot into a chatbot capable of mitigating negative emotions by pipelined two models, KR-BERT and KoGPT2-chatbot. We applied a creative method of giving emotions to unrefined everyday datasets through KR-BERT, and learning additional datasets through KoGPT2-chatbot. The development background of BERGPT-chatbot is as follows. Currently, the number of people with depression is increasing all over the world. This phenomenon is emerging as a more serious problem due to COVID-19, which causes people to increase long-term indoor living or limit interpersonal relationships. Overseas artificial intelligence chatbots aimed at relieving negative emotions or taking care of mental health care, have increased in use due to the pandemic. In Korea, Psychological diagnosis chatbots similar to those of overseas cases are being operated. However, as the domestic chatbot is a system that outputs a button-based answer rather than a text input-based answer, when compared to overseas chatbots, domestic chatbots remain at a low level of diagnosing human psychology. Therefore, we proposed a chatbot that helps mitigating negative emotions through BERGPT-chatbot. Finally, we compared BERGPT-chatbot and KoGPT2-chatbot through 'Perplexity', an internal evaluation metric for evaluating language models, and showed the superity of BERGPT-chatbot.

• Journal of Construction Engineering and Project Management
• /
• v.8 no.4
• /
• pp.1-24
• /
• 2018

In the last decade, the use of Building Information Modeling (BIM) as a new technology has been applied with traditional Computer-aided design implementations in an increasing number of architecture, engineering, and construction projects and applications. Its employment alongside construction management, can be a valuable tool in helping move these activities and projects forward in a more efficient and time-effective manner. The traditional stakeholders, i.e., Owner, A/E and the Contractor are involved in this BIM system that is used in almost every activity of construction projects, such as design, cost estimate and scheduling. This article extracts major features of the application of BIM from perspective of participating BIM components, along with the different phrases, and applies to them a logistic analysis using a fuzzy performance tree, quantifying these phrases to judge the effectiveness of the BIM techniques employed. That is to say, these fuzzy performance trees with fuzzy logic concepts can properly translate the linguistic rating into numeric expressions, and are thus employed in evaluating the influence of BIM applications as a mathematical process. The rotational fuzzy models are used to represent the membership functions of the performance values and their corresponding weights. Illustrations of the use of this fuzzy BIM performance tree are presented in the study for the uninitiated users. The results of these processes are an evaluation of BIM project performance as highly positive. The quantification of the performance ratings for the individual factors is a significant contributor to this assessment, capable of parsing vernacular language into numerical data for a more accurate and precise use in performance analysis. It is hoped that fuzzy performance trees and fuzzy set analysis can be used as a tool for the quality and risk analysis for other construction techniques in the future. Baldwin's rotational models are used to represent the membership functions of the fuzzy sets. Three scenarios are presented using fuzzy MEAN, AND and OR gates from the lowest to intermediate levels of the tree, and fuzzy SUM gate to relate the intermediate level to the top component of the tree, i.e., BIM application final performance. The use of fuzzy MEAN for lower levels and fuzzy SUM gates to reach the top level suggests the most realistic and accurate results. The methodology (fuzzy performance tree) described in this paper is appropriate to implement in today's construction industry when limited objective data is presented and it is heavily relied on experts' subjective judgment.

• Korean Chemical Engineering Research
• /
• v.57 no.5
• /
• pp.628-636
• /
• 2019

In various industrial processes, piping serves as a link between unit processes and is an essential installation for internal flow. Therefore, the optimum design of the piping system is very important in terms of safety and cost, which requires the estimation of the pressure drop, flow rate, pipe size, etc. in the piping system. In this study, we developed a software that determines pressure drop, flow rate, and pipe size when any two of these design variables are known. We categorized the flows into single phase, homogeneous two phase, and separated two phase flows, and applied suitable calculation models accordingly. We also constructed a system library for the calculation of the pipe material, relative roughness, fluid property, and friction coefficients to minimize user input. We further created a costing library according to the piping material for the calculation of the investment cost of the pipe per unit length. We implemented all these functions in an integrated environment using a graphical user interface for user convenience, and C # programming language. Finally, we verified the accuracy of the software using literature data and examples from an industrial process with obtained deviations of 1% and 8.8% for the single phase and two-phase models.

• Journal of Korean Medicine Rehabilitation
• /
• v.32 no.2
• /
• pp.65-81
• /
• 2022

Objectives This study aimed to assess the effects of oral administration of deer antler extracts on an osteoporosis-induced animal model. We analyzed the results of using deer antler single extracts on animal models with osteoporosis through a systematic review and meta-analysis. Methods We included osteoporosis studies in animal experiments that administrated deer antler extracts orally. We searched the following 13 databases without a language restriction: PubMed, EMBASE, Cochrane Library, Cumulative Index to Nursing and Allied Health Literature (CINAHL), China National Knowledge Infrastructure (CNKI), Wanfang, Korean Medical Database (KMbase), National Digital Science Library (NDSL), Korean Traditional Knowledge (Koreantk), Oriental Medicine Advanced Searching Integrated System (OASIS), Research Information Sharing Service (RISS), Korea Institute of Science and Technology Information (KISTI), and Koreanstudies Information Service System (KISS). We used Systematic Review Centre for Laboratory Animal Experimentation's risk of bias tool for assessing the methodological quality of the included studies. Results A total of 299 potentially relevant studies were searched and 11 were included for a systematic review. Nine studies used a single deer antler extract. A study compared the effects of single extracts of deer antler and antler glue, while another study compared the effects of three single extracts of deer antler, old antler, and antler glue. For evaluating the intervention effect, bone mineral density (BMD) was measured as the primary outcome, while the histomorphometric indicators of the bone and serum alkaline phosphatase and osteocalcin levels were used as the secondary outcome variables. On conducting a meta-analysis of studies on single deer antler extract, BMD was observed to be significantly increased compared to that in control group (standardized mean difference [SMD]=2.11; 95% confidence interval [CI]=1.58~2.65; Z=7.75; p<0.00001; I²=56%). As a result of meta-analysis, according to the concentration of deer antler, the group with high concentration showed statistically significantly higher BMD than the group with low concentration (SMD=1.28; 95% CI=0.74~1.82; Z=4.63; p<0.00001; I²=9%). Conclusions The research shows that the deer antler extracts have significant anti-osteoporotic effects on the osteoporosis-induced animal model. However the studies included in this research had a high methodological risk of bias. This indicates the requirement of considerable attention in the interpretation of the study results.