KJC Korea
Journal Central

In this paper, we propose Dynamic Rank Subsetting (DRAS) technique that enhances the energy-efficiency and the performance of memory system through the data compression. The goal of this technique is to enable a partial chip access by storing data in a compressed format within a subset of DRAM chips. To this end, a memory rank is dynamically configured to two independent sub-ranks. When writing a data block, it is compressed with a data compression algorithm and stored in one of the two sub-ranks. To service a memory request for the compressed data, only a sub-rank is accessed, whereas, for a memory request for the uncompressed data, two sub-ranks are accessed as done in the conventional memory systems. Since DRAS technique requires minimal hardware modification, it can be used in the conventional memory systems with low hardware overheads. Through experimental evaluation with a memory simulator, we show that the proposed technique improves the performance of the memory system by 12% on average and reduces the power consumption of memory system by 24% on average.

In this paper, we propose a method to build a knowledge base based on natural language processing for intelligent battlefield awareness service. The current command and control system manages and utilizes the collected battlefield information and tactical data at a basic level such as registration, storage, and sharing, and information fusion and situation analysis by an analyst is performed. This is an analyst's temporal constraints and cognitive limitations, and generally only one interpretation is drawn, and biased thinking can be reflected. Therefore, it is essential to aware the battlefield situation of the command and control system and to establish the intellignet decision support system. To do this, it is necessary to build a knowledge base specialized in the command and control system and develop intelligent battlefield awareness services based on it. In this paper, among the entity names suggested in the exobrain corpus, which is the private data, the top 250 types of meaningful names were applied and the weapon system entity type was additionally identified to properly represent battlefield information. Based on this, we proposed a way to build a battlefield-aware knowledge base through mention extraction, cross-reference resolution, and relationship extraction.

In this paper, we propose a machine learning method for diagnosing the failure of a gas pressure regulator. Originally, when implementing a machine learning model for detecting abnormal operation of a facility, it is common to install sensors to collect data. However, failure of a gas pressure regulator can lead to fatal safety problems, so that installing an additional sensor on a gas pressure regulator is not simple. In this paper, we propose various machine learning approach for diagnosing the abnormal operation of a gas pressure regulator with only the flow rate and gas pressure data collected from a gas pressure regulator itself. Since the fault data of a gas pressure regulator is not enough, the model is trained in all classes by applying the over-sampling method. The classification model was implemented using Gradient boosting, 1D Convolutional Neural Networks, and LSTM algorithm, and gradient boosting model showed the best performance among classification models with 99.975% accuracy.

In this paper, we propose a new structured entity recognition DeNERT model. Recently, the field of natural language processing has been actively researched using pre-trained language representation models with a large amount of corpus. In particular, the named entity recognition, which is one of the fields of natural language processing, uses a supervised learning method, which requires a large amount of training dataset and computation. Reinforcement learning is a method that learns through trial and error experience without initial data and is closer to the process of human learning than other machine learning methodologies and is not much applied to the field of natural language processing yet. It is often used in simulation environments such as Atari games and AlphaGo. BERT is a general-purpose language model developed by Google that is pre-trained on large corpus and computational quantities. Recently, it is a language model that shows high performance in the field of natural language processing research and shows high accuracy in many downstream tasks of natural language processing. In this paper, we propose a new named entity recognition DeNERT model using two deep learning models, DQN and BERT. The proposed model is trained by creating a learning environment of reinforcement learning model based on language expression which is the advantage of the general language model. The DeNERT model trained in this way is a faster inference time and higher performance model with a small amount of training dataset. Also, we validate the performance of our model's named entity recognition performance through experiments.

The development process of deep learning is an iterative task that requires a lot of manual work. Among the steps in the development process, pre-processing of learning data is a very costly task, and is a step that significantly affects the learning results. In the early days of AI's algorithm research, learning data in the form of public DB provided mainly by data scientists were used. The learning data collected in the real environment is mostly the operational data of the sensors and inevitably contains various noises. Accordingly, various data cleaning frameworks and methods for removing noises have been studied. In this paper, we proposed a method for detecting and removing noises from time-series data, such as sensor data, that can occur in the IoT environment. In this method, the linear regression method is used so that the system repeatedly finds noises and provides data that can replace them to clean the learning data. In order to verify the effectiveness of the proposed method, a simulation method was proposed, and a method of determining factors for obtaining optimal cleaning results was proposed.

Blockchain is a distributed shared ledger that transparently manages information through verification and agreement between nodes connected to a distributed network. Recently, cases of data management among authorized agencies based on private blockchain are increasing. In this paper, we investigated the application cases and technical processes of PBFT, the representative consensus algorithm of private blockchain, and proposed a modified PBFT algorithm that enables efficient consensus by simplifying duplicate verification and consensus processes that occur during PBFT processing. The algorithm proposed in this paper goes through the process of selecting a delegation node through an authoritative node and can increase the safety of the delegation node selection process by considering an efficient re-election algorithm for candidate nodes. By utilizing this research, it is possible to reduce the burden on the network communication cost of the consensus process and effectively process the final consensus process between nodes. ▸Key

In this paper, we present a novel approach to efficiently control the location of NPC(Non-playable characters) in the interactive virtual world such as game, virtual reality. To control the NPC's movement path, we first calculate the main trajectory based on the user's path, and then move the NPC based on the weight map. Our method constructs automatically a navigation mesh that provides new paths for NPC by referencing the user trajectories. Our method enables adaptive changes to the virtual world over time and provides user-preferred path weights for smartagent path planning. We have tested the usefulness of our algorithm with several example scenarios from interactive worlds such as video games, virtual reality. In practice, our framework can be applied easily to any type of navigation in an interactive world.

In this paper, we propose a globally successful animation rhythm type. Based on the study of spatial variables in Campbell's heroic narrative and the study on the duration of shots in the video, the rhythm of the shot according to the spatial narrative of <Frozen(2013)> was analyzed. Through this, three conclusions were drawn. First, the basis for the visualization of narrative intensity is possible through the shot density of <Frozen> narrative, Second, a shot rhythm type was presented, which was presented as an ascending type, a descending type, a mountain type, a depressed type, and a complex type, and the characteristics of the narrative were analyzed. Third, the strength of the shot rhythm shown in the hero narrative spatial map model was divided into top, middle, and bottom, and the measurement criteria of narrative strength were presented. This study is meaningful in that it visualized and typified artistic emotions as objective data in the flow of animation content focused on philosophical and qualitative methods.

Through the consensus algorithm, which is the core technology of the blockchain, the same data is accurately shared between connected nodes. The use of an appropriate consensus algorithm that considers the user and the usage environment ensures efficient maintenance of data integrity and accuracy. In this paper, we proposed a performance evaluation method for efficient selection of a consensus algorithm among authorized nodes considering the characteristics of a private blockchain platform, and applied the modified item to the existing published formula considering the number of authoritative connected nodes. Through this process, it was possible to simplify the consensus process considering the stability between nodes. The stability of the consensus process can be improved by selecting an appropriate consensus algorithm based on the proposed research.

In this paper, we propose an efficient emotional relaxation framework using Dijkstra algorithm based on anisotropic features. Emotional relaxation is as important as emotional analysis. This is a framework that can automatically alleviate the person's depression or loneliness. This is very important for HCI (Human-Computer Interaction). In this paper, 1) Emotion value changing from facial expression is calculated using Microsoft 's Emotion API, 2) Using these differences in emotion values, we recognize abnormal feelings such as depression or loneliness. 3) Finally, emotional mesh based matching process considering the emotional histogram and anisotropic characteristics is proposed, which suggests emotional relaxation to the user. In this paper, we propose a system which can recognize the change of emotion easily by using face image and train personal emotion by emotion relaxation.

In this paper, we examine the problems of the current Internet due to the development of network services and the expansion of network bandwidth. The current Internet has been used for a long time because it is composed of TCP / IP, but fundamental problems such as bandwidth, transmission rate, and security have not been solved. Therefore, the future network must be prepared through continuous investment and maintenance. In order to overcome this problem, we will propose a way to overcome the above problems and upgrade by converting the current Internet Protocol into the next generation network. Currently, many researches on next-generation networks have been conducted, but there are not many studies in Korea, and research on next-generation networks will be a very important task for the future development of the Internet service industry at the national level. In this paper, we propose an advanced internet environment through the advantages of various next generation protocols.

Wireless sensor networks collect and analyze sensing data in a variety of environments without human intervention. The sensor network changes its lifetime depending on routing protocols initially installed. In addition, it is difficult to modify the routing path during operating the network because sensors must consume a lot of energy resource. It is important to measure the network performance through simulation before building the sensor network into the real field. This paper proposes a WSN model for a low-energy adaptive clustering hierarchy protocol using DEVS kernel models. The proposed model is implemented with the sub models (i.e. broadcast model and controlled model) of the kernel model. Experimental results indicate that the broadcast model based WSN model showed lower CPU resource usage and higher message delivery than the broadcast model.

In the fourth industrial revolution represented by hyper-connected and intelligence, cloud computing is drawing attention as a technology to realize big data and artificial intelligence technologies. The proliferation of cloud computing has also increased the number of threats. In this paper, we propose one way to effectively monitor to the resources assigned to clients by the IaaS service provider. The method we propose in this paper is to model the use of resources allocated to cloud systems using ARIMA algorithm, and it identifies abnormal situations through the use and trend analysis. Through experiments, we have verified that the client service provider can effectively monitor using the proposed method within the minimum amount of access to the client systems.

With the rapid development of information and communication technology (ICT), various sensors are being embedded in wearable devices. Consequently, these devices can continuously collect data including health data from individuals. The collected health data can be used not only for healthcare services but also for analyzing an individual’s lifestyle by combining with other external data. This helps in making an individual’s life more convenient and healthier. However, collecting health data may lead to privacy issues since the data is personal, and can reveal sensitive insights about the individual. Thus, in this paper, we present a method to collect an individual’s health data from a smart band in a privacy-preserving manner. We leverage the local differential privacy to achieve our goal. Additionally, we propose a way to find feature points from health data. This allows for an effective trade-off between the degree of privacy and accuracy. We carry out experiments to demonstrate the effectiveness of our proposed approach and the results show that, with the proposed method, the error rate can be reduced upto 77%.

The pattern query in graph database has advantages of easy query expression and high query processing performance compared to relational database SQL. However, unlike the relational database, the graph database may not utilize the advantages of pattern query depending on modeling because the methodology for building the logical data model is not defined. In this study, in the is-a node modeling method that appears during the graph modeling process, we experiment that there is a difference in performance between graph pattern query when designing with a generalization model and designing with a specialization model. As a result of the experiment, it was shown that better performance can be obtained when the is-a node is designed as a specialization model. In addition, when writing a pattern query, we show that if a variable is bound to a node or edge, performance may be better than that of the variable of not bounded. The experimental results can be presented as an is-a node modeling method for pattern query and a graph query writing method in the graph database.

Emotional analysis is an application of opinion mining that analyzes opinions and tendencies of people appearing in unstructured text. Recently, emotional analysis of social media has attracted attention, but social media contains newly-created words and slang, so it is not easy to analyze with existing emotional analysis. In this study, I design a new emotional analysis system to solve these problems. The proposed system is possible to analyze various emotions as well as positive and negative in social media including newly-created words and slang. First, I collect newly-created words and slang related to emotions that appear in social media. Then, expand the existing emotional model and use it to quantify the degree of sentiment in emotional words. Also, a new sentiment dictionary is constructed by reflecting the degree of sentiment. Finally, I design an emotional analysis system that applies an sentiment dictionary that includes newly-created words and an extended emotional model.

Automatic term extraction is to recognize domain-specific terms given a collection of domain-specific text. Previous term extraction methods operate effectively in unsupervised manners which include extracting candidate terms, and assigning importance scores to candidate terms. Regarding the calculation of term importance scores, the study focuses on utilizing sets of inner and outer terms of a candidate term. For a candidate term, its inner terms are shorter terms which belong to the candidate term as components, and its outer terms are longer terms which include the candidate term as their component. This work presents various functions that compute, for a candidate term, term strength from either set of its inner or outer terms. In addition, a scoring method of a term importance is devised based on C-value score and the term strength values obtained from the sets of inner and outer terms. Experimental evaluations using GENIA and ACL RD-TEC 2.0 datasets compare and analyze the effectiveness of the proposed term extraction methods for English. The proposed method performed better than the baseline method by up to 1% and 3% respectively for GENIA and ACL datasets.

In the Fourth Industrial Revolution, successful cases using big data in various industries are reported. This paper examines cases that successfully use big data in the medical industry to develop the service and draws implications in value that big data create. The related work introduces big data technology in the medical field and cases of eight innovative service in the big data service are explained. In the introduction, the overall structure of the study is mentioned by describing the background and direction of this study. In the literature study, we explain the definition and concept of big data, and the use of big data in the medical industry. Next, this study describes the several cases, such as technologies using national health information and personal genetic information for the study of diseases, personal health services using personal biometric information, use of medical data for efficiency of business processes, and medical big data for the development of new medicines. In the conclusion, we intend to provide direction for the academic and business implications of this study, as well as how the results of the study can help the domestic medical industry.

The purpose of this study is to explore the impact of Union-Pay mobile payments on user viscosity from the perspective of service quality and brand personality. In the future, it is meaningful for CUP to obtain a stable and loyal user group in China's mobile payment market. This study uses SPSS22.0 and AMOS statistical analysis tools to conduct empirical research. In this case, this study uses mobile service quality and The brand personality is the independent variable, and the user's viscosity is the dependent variable, which studies the impact on the user's viscosity in the context of China Union-Pay mobile payment. the study. According to the analysis results, the research goal: service quality has a significant positive correlation effect on user perceived value; service quality has a significant positive correlation effect on user viscosity; perceived value has significant positive correlation effect on user viscosity; brand personality to user Viscosity has a significant positive correlation effect; brand personality has a significant positive correlation effect on user perceived value. Through this research, we can make Suggestions on the industrial development of mobile payment enterprises and provide better opinions on the development of mobile payment.

In this paper we propose an uses on-device-based edge computing technology and big data analysis methods through the use of on-device-based edge computing technology and analysis of big data, which are distributed computing paradigms that introduce computations and storage devices where necessary to solve problems such as transmission delays that occur when data is transmitted to central centers and processed in current general smart factories. However, even if edge computing-based technology is applied in practice, the increase in devices on the network edge will result in large amounts of data being transferred to the data center, resulting in the network band reaching its limits, which, despite the improvement of network technology, does not guarantee acceptable transfer speeds and response times, which are critical requirements for many applications. It provides the basis for developing into an AI-based facility prediction conservation analysis tool that can apply deep learning suitable for big data in the future by supporting intelligent facility management that can support productivity growth through research that can be applied to the field of facility preservation and smart factory industry with integrated hardware technology that can accommodate these requirements and factory management and control technology.

The objective of the present study was to identify the factors influencing the smartphone addiction among college students and to obtain basic information in terms of enhancing proper use of the smartphone. Based on the previous literature with constructs of depression, aggression, self-control and smartphone addiction, a research model was prepared. A total of 332 students were selected from 5 universities in G Metropolitan City and collected data were analyzed through hierarchical multiple regression. The major results of the study were as follows. First, self-control was the most powerful predictor of smartphone addiction. Second, anger caused smartphone addiction to increase whereas social experience caused smartphone addiction to decrease. Third, depression did not cause smartphone addiction to increase. Finally, implications for preventing and decreasing the smartphone addiction among college students were also provided.

This study is primarily aimed at providing a foundation for academic development and the leap forward of the Special Security Industry through the press report analysis on Korea's special security guard using big data. The research methods It was analyzed by the research methods in relation to keyword trends for 'special security guard’ and 'special guards' using the Big Kinds program. According to the analysis based on the period of growth (quantitative and qualitative) of the special security industry, there were many press reports and exposure related to carrying firearms, national major facilities, and regular employees. Unlike the general security guards, the special security guards were released higher by media as a law was revised to allow them to carry or use firearms at important national facilities. There was a lot of media attention concerned about the side effects of misuse, and there were also high media reports on the transition of regular workers to improve poor treatment, such as the unstable status of special security guards and low wages. Therefore, the need for continuous development and improvement of professionalism and work efficiency of special security services are emphasized.

The whole world is anxious that aircraft could be used as a tool for terrorism after 9/11. The disappearance of Malaysia Airlines is again adding to fears about aircraft. Because these aircraft attacks cause many human casualties, the purpose of the study is to analyze the problems in the current air security system and to present alternatives. The methodology of the study used a literature research methodology to review the current status of aircraft terrorism and related regulations, such as current aviation-related laws and anti-terrorism laws, and prior studies. The purpose of the Chapter is to present an aviation security system that promotes the safety of air traffic through the prevention of aircraft terror by presenting the roles and improvement measures of aviation security personnel, foreign police officers, intelligence agencies, and legal blind spots and flaws.

The purpose of this study is to examine the influence of social support perceived by multicultural youth on learning adaptation and to identify the indirect effects of self-esteem and achievement motivation in this relationship. In order to analyze these relationships through structural equation model analysis, panel data were used for the seventh year of multicultural youth (2017). As a result of the analysis, first, the social support, self-esteem and achievement motivation perceived by multicultural youth had a significant influence on learning adaptation. Second, self-esteem of multicultural youths had a significant effect on social support. Third, the influence of social support perceived by multicultural youth on learning adaptation was found to have indirect effects through self-esteem and achievement motivation as well as direct effects. Based on these findings, this study suggested practical intervention plans to improve social support and learning adaptation of multicultural youths.

We propose a empirically analyzed the relationship between organizational citizenship behavior, empowerment and job satisfaction of 546 office workers in general companies in Seoul city using SPSS 23.0 and AMOS 21.0. The results of this study were as follows: First, participatory behavior positively influenced self-determination in the relationship between empowerment and organizational citizenship behavior. Second, altruism had a positive effect on job satisfaction in the relationship between organizational citizenship behavior and job satisfaction. Third, looking at the aspect of direct and indirect effects on job satisfaction, it was noticeable that the altruism of organizational citizenship behavior had a higher direct effect, and that participatory behavior had a higher figure of the indirect effect. On the other hand, self-determination and influence/meaning, which are factors of empowerment, showed a higher figure than organizational citizenship behavior. Therefore, it is noticeable that empowerment is the most important factor in terms of affecting job satisfaction. that is, continuous monitoring of the effectiveness of the empowerment granted to members and improvement of the meaning and self-determination of the role at the workplace through training will increase the autonomy and it will contribute to the activation of the organization.

Recently, as awareness of the need for software education has spread worldwide, the government of Korea has led compulsory software education also. Basic software education in universities has been stabilized through various trials and efforts. However, due to software classes are mandatory, students not only could not have motivation for learning but also have treated programming course as a difficult subject. In this paper, two programming classes, which were designed and managed as a problem-oriented programming class for the purpose of cultivating computational thinking for the non-computer science students, are compared using the lecture assessment results. As a result, in the case of expanding the use of the problem as a grammatical explanation aid and expanding the ratio of major-friendly problems, the student's responses were concentrated on higher scores and the response average improved by about 7%. It means that the level of difficulty experienced by learners is lowered.