• KEPCO Journal on Electric Power and Energy
• /
• v.6 no.4
• /
• pp.413-418
• /
• 2020

It is necessary to manage the prediction accuracy of the machine learning model to prevent the decrease in the performance of the grid network condition prediction model due to overfitting of the initial training data and to continuously utilize the prediction model in the field by maintaining the prediction accuracy. In this paper, we propose an automation technique for maintaining the performance of the model, which increases the accuracy and reliability of the prediction model by considering the characteristics of the power grid state data that constantly changes due to various factors, and enables quality maintenance at a level applicable to the field. The proposed technique modeled a series of tasks for maintaining the performance of the power grid condition prediction model through the application of the workflow management technology in the form of a workflow, and then automated it to make the work more efficient. In addition, the reliability of the performance result is secured by evaluating the performance of the prediction model taking into account both the degree of change in the statistical characteristics of the data and the level of generalization of the prediction, which has not been attempted in the existing technology. Through this, the accuracy of the prediction model is maintained at a certain level, and further new development of predictive models with excellent performance is possible. As a result, the proposed technique not only solves the problem of performance degradation of the predictive model, but also improves the field utilization of the condition prediction model in a complex power grid system.

• The Journal of Bigdata
• /
• v.3 no.2
• /
• pp.59-70
• /
• 2018

In order to provide stable district heat supplying service to the certain limited residential area, it is the most important to forecast the short-term future demand more accurately and produce and supply heat in efficient way. However, it is very difficult to develop a universal heat demand forecasting model that can be applied to general situations because the factors affecting the heat consumption are very diverse and the consumption patterns are changed according to individual consumers and regional characteristics. In particular, considering all of the various variables that can affect heat demand does not help improve performance in terms of accuracy and versatility. Therefore, this study aims to develop a demand forecasting model using deep learning based on only limited information that can be acquired in real time. A demand forecasting model was developed by learning the artificial neural network of the Tensorflow using past data consisting only of the outdoor temperature of the area and date as input variables. The performance of the proposed model was evaluated by comparing the accuracy of demand predicted with the previous regression model. The proposed heat demand forecasting model in this research showed that it is possible to enhance the accuracy using only limited variables which can be secured in real time. For the demand forecasting in a certain region, the proposed model can be customized by adding some features which can reflect the regional characteristics.

• Journal of Internet Computing and Services
• /
• v.22 no.4
• /
• pp.1-11
• /
• 2021

As the Internet developed, various and complex cyber attacks began to emerge. Various detection systems were used outside the network to defend against attacks, but systems and studies to detect attackers inside were remarkably rare, causing great problems because they could not detect attackers inside. To solve this problem, studies on the lateral movement detection system that tracks and detects the attacker's movements have begun to emerge. Especially, the method of using the Remote Desktop Protocol (RDP) is simple but shows very good results. Nevertheless, previous studies did not consider the effects and relationships of each logon host itself, and the features presented also provided very low results in some models. There was also a problem that the model could not explain why it predicts that way, which resulted in reliability and robustness problems of the model. To address this problem, this study proposes an interpretable RDP-based lateral movement detection system using page rank algorithm and SHAP(Shapley Additive Explanations). Using page rank algorithms and various statistical techniques, we create features that can be used in various models and we provide explanations for model prediction using SHAP. In this study, we generated features that show higher performance in most models than previous studies and explained them using SHAP.

• KIPS Transactions on Software and Data Engineering
• /
• v.10 no.1
• /
• pp.1-8
• /
• 2021

In the political field of news articles, there are polarized and biased characteristics such as conservative and liberal, which is called political bias. We constructed keyword-based dataset to classify bias of news articles. Most embedding researches represent a sentence with sequence of morphemes. In our work, we expect that the number of unknown tokens will be reduced if the sentences are constituted by subwords that are segmented by the language model. We propose a document embedding model with subword tokenization and apply this model to SVM and feedforward neural network structure to classify the political bias. As a result of comparing the performance of the document embedding model with morphological analysis, the document embedding model with subwords showed the highest accuracy at 78.22%. It was confirmed that the number of unknown tokens was reduced by subword tokenization. Using the best performance embedding model in our bias classification task, we extract the keywords based on politicians. The bias of keywords was verified by the average similarity with the vector of politicians from each political tendency.

• Journal of the Korean Society of Radiology
• /
• v.16 no.1
• /
• pp.1-6
• /
• 2022

Deep learning is a collection of algorithms that enable learning by summarizing the key contents of large amounts of data; it is being developed to diagnose lesions in the medical imaging field. To evaluate the accuracy of the cerebral hemorrhage diagnosis, we used a convolutional neural network (CNN) to derive the diagnostic accuracy of cerebral parenchyma computed tomography (CT) images and the cerebral parenchyma CT images of areas where cerebral hemorrhages are suspected of having occurred. We compared the accuracy of CNN with different numbers of hidden layers and discovered that CNN with more hidden layers resulted in higher accuracy. The analysis results of the derived CT images used in this study to determine the presence of cerebral hemorrhages are expected to be used as foundation data in studies related to the application of artificial intelligence in the medical imaging industry.

• Knowledge Management Research
• /
• v.23 no.2
• /
• pp.193-210
• /
• 2022

The rapid development of mobile technology and the improvement of network speed are providing convenience to various services, and mobile shopping malls are no exception. Although efforts are being made to promote sales by combining various technologies such as customized recommendations using big data and specialized personalization services based on artificial intelligence, most mobile shopping malls have the same detailed page information structure including detailed product information. In this context, in this study, it was determined that the content layout of the product detail page and the mobile product detail page layout tailored to the consumer's preference should be presented according to the consumer's preference. Based on Higgins' Regulatory Focus Theory, a study of consumer propensity revealed that the content layout arrangement on a product detail page, when presented in an F-shape, informs the consumer that it is organized. If presented in a Z-shape, vivid information was recognized, and it was examined whether the product attitude and purchase intention were affected. As a result, when the content layout composition was presented as a layout arrangement in the form of a sense of unity and organization, prevention-focused consumers were positively affected by product attitudes and purchase intentions, and promotion-oriented consumers felt freedom. When presented in an arrangement, it was confirmed that the product attitude and purchase intention were affected.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.23 no.4
• /
• pp.189-196
• /
• 2023

As the manufacturing paradigm shifts, various collaborative robots are creating new markets. Demand for collaborative robots is increasing in all industries for the purpose of easy operation, productivity improvement, and replacement of manpower who do simple tasks compared to existing industrial robots. However, accidents frequently occur during work caused by collaborative robots in industrial sites, threatening the safety of workers. In order to construct an industrial site through robots in a human-centered environment, the safety of workers must be guaranteed, and there is a need to develop a collaborative robot guard system that provides reliable communication without the possibility of dispatch. It is necessary to double prevent accidents that occur within the working radius of cobots and reduce the risk of safety accidents through sensors and computer vision. We build a system based on OPC UA, an international protocol for communication with various industrial equipment, and propose a collaborative robot guard system through image analysis using ultrasonic sensors and CNN (Convolution Neural Network). The proposed system evaluates the possibility of robot control in an unsafe situation for a worker.

• Journal of Broadcast Engineering
• /
• v.27 no.1
• /
• pp.104-123
• /
• 2022

Style transfer based on neural network provides very high quality results by reflecting the high level structural characteristics of images, and thereby has recently attracted great attention. This paper deals with the problem of resolution limitation due to GPU memory in performing such neural style transfer. We can expect that the gradient operation for style transfer based on partial image, with the aid of the fixed size of receptive field, can produce the same result as the gradient operation using the entire image. Based on this idea, each component of the style transfer loss function is analyzed in this paper to obtain the necessary conditions for partitioning and padding, and to identify, among the information required for gradient calculation, the one that depends on the entire input. By structuring such information for using it as auxiliary constant input for partition-based gradient calculation, this paper develops a recursive algorithm for super high-resolution image style transfer. Since the proposed method performs style transfer by partitioning input image into the size that a GPU can handle, it can perform style transfer without the limit of the input image resolution accompanied by the GPU memory size. With the aid of such super high-resolution support, the proposed method can provide a unique style characteristics of detailed area which can only be appreciated in super high-resolution style transfer.

• KIPS Transactions on Software and Data Engineering
• /
• v.12 no.7
• /
• pp.303-314
• /
• 2023

Voice conversion, a technology that allows an individual's speech data to be regenerated with the acoustic properties(tone, cadence, gender) of another, has countless applications in education, communication, and entertainment. This paper proposes an approach based on the StarGAN-VC model that generates realistic-sounding speech without requiring parallel utterances. To overcome the constraints of the existing StarGAN-VC model that utilizes one-hot vectors of original and target speaker information, this paper extracts feature vectors of target speakers using a pre-trained version of Rawnet3. This results in a latent space where voice conversion can be performed without direct speaker-to-speaker mappings, enabling an any-to-any structure. In addition to the loss terms used in the original StarGAN-VC model, Wasserstein distance is used as a loss term to ensure that generated voice segments match the acoustic properties of the target voice. Two Time-Scale Update Rule (TTUR) is also used to facilitate stable training. Experimental results show that the proposed method outperforms previous methods, including the StarGAN-VC network on which it was based.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.05a
• /
• pp.171-173
• /
• 2022

The 4th industrial revolution technology is developing people's lives more efficiently. GIS provided on the Internet services such as traffic information and time information makes people getting more quickly to destination. National geographic information service(NGIS) and each local government are making basic data to investigate SOC accessibility for analyzing optimal point. To construct the shortest distance, the accessibility from the starting point to the arrival point is analyzed. Applying road network map, the starting point and the ending point, the shortest distance, the optimal accessibility is calculated by using Dijkstra algorithm. The analysis information from multiple starting points to multiple destinations was required more than 3 steps of manual analysis to decide the position for the optimal point, within about 0.1% error. It took more time to process the many-to-many (M×N) calculation, requiring at least 32G memory specification of the computer. If an optimal proximity analysis service is provided at a desired location more versatile, it is possible to efficiently analyze locations that are vulnerable to business start-up and living facilities access, and facility selection for the public.