• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.4
• /
• pp.169-178
• /
• 2022

Morphemes are most primitive units in a language that lose their original meaning when segmented into smaller parts. In Korean, a sentence is a sequence of eojeols (words) separated by spaces. Each eojeol comprises one or more morphemes. Korean morphological analysis (KMA) is to divide eojeols in a given Korean sentence into morpheme units. It also includes assigning appropriate part-of-speech(POS) tags to the resulting morphemes. KMA is one of the most important tasks in Korean natural language processing (NLP). Improving the performance of KMA is closely related to increasing performance of Korean NLP tasks. Recent research on KMA has begun to adopt the approach of machine translation (MT) models. MT is to convert a sequence (sentence) of units of one domain into a sequence (sentence) of units of another domain. Neural machine translation (NMT) stands for the approaches of MT that exploit neural network models. From a perspective of MT, KMA is to transform an input sequence of units belonging to the eojeol domain into a sequence of units in the morpheme domain. In this paper, we propose a deep learning model for KMA. The backbone of our model is based on the BERT-fused model which was shown to achieve high performance on NMT. The BERT-fused model utilizes Transformer, a representative model employed by NMT, and BERT which is a language representation model that has enabled a significant advance in NLP. The experimental results show that our model achieves 98.24 F1-Score.

• Journal of Intelligence and Information Systems
• /
• v.28 no.2
• /
• pp.263-278
• /
• 2022

With the recent advent of IoT technology, automatic pet feeders are being distributed so that owners can feed their companion animals while they are out. However, due to behaviors of pets, the method of measuring weight, which is important in automatic feeding, can be easily damaged and broken when using the scale. The 3D camera method has disadvantages due to its cost, and the 2D camera method has relatively poor accuracy when compared to 3D camera method. Hence, the purpose of this study is to propose a deep learning approach that can accurately estimate weight while simply using a 2D camera. For this, various convolutional neural networks were used, and among them, the ResNet101-based model showed the best performance: an average absolute error of 3.06 grams and an average absolute ratio error of 3.40%, which could be used commercially in terms of technical and financial viability. The result of this study can be useful for the practitioners to predict the weight of a standardized object such as feed only through an easy 2D image.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.28 no.2
• /
• pp.385-395
• /
• 2018

With the development of Internet, cyber attack has become a major threat. To detect cyber attacks, intrusion detection system(IDS) has been widely deployed. But IDS has a critical weakness which is that it generates a large number of false alarms. One of the promising techniques that reduce the false alarms in real time is machine learning. However, there are problems that must be solved to use machine learning. So, many machine learning approaches have been applied to this field. But so far, researchers have not focused on features. Despite the features of IDS alerts are important for performance of model, the approach to feature is ignored. In this paper, we propose new feature set which can improve the performance of model and can be extracted from a single alarm. New features are motivated from security analyst's know-how. We trained and tested the proposed model applied new feature set with real IDS alerts. Experimental results indicate the proposed model can achieve better accuracy and false positive rate than SVM model with ordinary features.

• Journal of Korean Society of Transportation
• /
• v.23 no.5 s.83
• /
• pp.157-167
• /
• 2005

Seoul metropolitan public transport reform results in the introduction of the semi-public operation and distance-based fare policies. With implementation of these policies, public transport revenue allocation has been (will be) evolved very complicated because the existing revenue allocation issues have not only been clearly solved, which is generated by the combined relationship among Korea Railroad Corporation (KRC). Seoul Metropolitan Subway Corporation (SMSC). Seoul Metropolitan Rapid Transit Corporation (SMRTC), and Incheon Rapid Transit Corporation (IRTC), but also the revenue allocation problem between bus and urban railroad-related organizations need to be considered in this combined framework. On top of that. based on the future plans such as the private sector's railroad construction plan(s), the light rail transit construction plans of several local governments and the join of remained bus lines of Seoul metropolitan areas, it is understood that the revenue allocation among public transport operating organization will become one of main issues of operation organization as well as local and central governments. As a basic approach for revenue allocation of public transport operation organizations, the purpose of this paper is to propose an integrated model applicable to estimate degree of service contribution in passenger carriage in the combined public transport network. With a hypothesis that the complete electronic card system is deployed, this paper supposes every passenger's loading and alighting stations is recordable. Thereby, this paper limits research scope as to Seoul metropolitan railroad area since used route(s) between origin and destination stations can not be traceded because transfer stations each passenger path through is not recorded. Each model proposed in the paper is as follows: 1. a generalized cost reflecting passenger's transfer behavior; 2.a K path model for determining similar routes between O-D; 3.an assignment model for loading O-D trips onto the detected similar routes using Logit Model.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.46 no.3
• /
• pp.48-58
• /
• 2009

Geo-graphical routing protocols as GPSR are known to be very suitable and useful for vehicular ad-hoc networks. However, a corresponding node can include some stale neighbor nodes being out of a transmission range, and the stale nodes are pone to get a high priority to be a next relay node in the greedy mode. In addition, some useful redundant information can be eliminated during planarization in the recovery mode. This paper deals with a new recovery mode, the Greedy Border Superiority Routing(GBSR), along with an Adaptive Neighbor list Management(ANM) scheme. Each node can easily treat stale nodes on its neighbor list by means of comparing previous and current Position of a neighbor. When a node meets the local maximum, it makes use of a border superior graph to recover from it. This approach improve the packet delivery ratio while it decreases the time to recover from the local maximum. We evaluate the performance of the proposed methods using a network simulator. The results shown that the proposed protocol reveals much better performance than GPSR protocol. Please Put the of paper here.

• Journal of Korea Water Resources Association
• /
• v.53 no.7
• /
• pp.493-505
• /
• 2020

The rainfall-runoff model has been generally adopted to obtain a consistent runoff sequence with the use of the long-term ground-gauged based precipitation data. The Thiessen polygon is a commonly applied approach for estimating the mean areal rainfall from the ground-gauged precipitation by assigning weight based on the relative areas delineated by a polygon. However, spatial bias is likely to increase due to a sparse network of the rain gauge. This study aims to generate continuous runoff sequences with the mean areal rainfall obtained from radar rainfall estimates through a PRMS rainfall-runoff model. Here, the systematic error of radar rainfall is corrected by applying the G/R Ratio. The results showed that the estimated runoff using the corrected radar rainfall estimates are largely similar and comparable to that of the Thiessen. More importantly, one can expect that the mean areal rainfall obtained from the radar rainfall estimates are more desirable than that of the ground in terms of representing rainfall patterns in space, which in turn leads to significant improvement in the estimation of runoff.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.25 no.1
• /
• pp.27-35
• /
• 2005

Ultrasonic inspection methods are widely used for detecting flaws in materials. The signal analysis step plays a crucial part in the data interpretation process. A number of signal processing methods have been proposed to classify ultrasonic flaw signals. One of the more popular methods involves the extraction of an appropriate set of features followed by the use of a neural network for the classification of the signals in the feature spare. This paper describes an alternative approach which uses the least mean square (LMS) method and exportation maximization (EM) algorithm with the model based deconvolution which is employed for classifying nondestructive evaluation (NDE) signals from steam generator tubes in a nuclear power plant. The signals due to cracks and deposits are not significantly different. These signals must be discriminated to prevent from happening a huge disaster such as contamination of water or explosion. A model based deconvolution has been described to facilitate comparison of classification results. The method uses the space alternating generalized expectation maximiBation (SAGE) algorithm ill conjunction with the Newton-Raphson method which uses the Hessian parameter resulting in fast convergence to estimate the time of flight and the distance between the tube wall and the ultrasonic sensor. Results using these schemes for the classification of ultrasonic signals from cracks and deposits within steam generator tubes are presented and showed a reasonable performances.

• Journal of Digital Convergence
• /
• v.18 no.4
• /
• pp.121-127
• /
• 2020

Innovation in local industries and the development of public services require, among other things, shared growth with the concept of regional co-prosperity. Regional co-prosperity is essentially aimed at bridging regional balance or regional gaps, which means a relationship that can achieve shared growth through complementary cooperation. In this study, cases using living labs were drawn based on the current status of industry-academic cooperation at the level of regional innovation and its policy implications were sought. Local governments are making various attempts to solve regional problems and enhance the linkage effect of securing mutual competitiveness through co-prosperity cooperation autonomously. In particular, an effective approach has been continued by activating the Living Lab Network, a problem-solving mechanism, focusing on pending regional issues. Above all, one of the strategies for regional development should be linked to the establishment of a cooperative system for win-win cooperation and policy means to support it. The activation of cooperative programs with local universities, companies and local governments and the case of problem-solving using living labs. Therefore, it suggested that active participation by various stakeholders and a cooperative governance model were needed to enable Living Lab.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.16 no.2
• /
• pp.215-221
• /
• 2006

This paper is concerned with the control of multiple nonlinearities included in a humanoid robot system. A humanoid robot has some problems such as the structural instability, which leads to consider the control of multiple nonlinearities caused by driver parts as well as gear reducer. Saturation and backlash are typical examples of nonlinearities in the system. The conventional algorithms of backlash control were fuzzy algorithm, disturbance observer and neural network, etc. However, it is not easy to control the system by employing only single algorithm since the system usually includes multiple nonlinearities. In this paper, a switching Pill is considered for a control of saturation and a dual feedback algorithm is proposed for a backlash control. To implement the above algorithms, the system identification is firstly performed for the minimization of the difference between the results of simulation and experiment, and then the switching Pill gains are determined using genetic algorithm with some heuristic approach. The performance of the switching Pill controller for saturation and the dual feedback for backlash control is investigated through the simulation. Finally, it is shown that the implemented control system has good results and can be applied to the real humanoid robot system ISHURO.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.36 no.4B
• /
• pp.332-345
• /
• 2011

As weapon systems are evolving into more advanced and complex ones, the role of the software is becoming heavily significant in their developments. Particularly in the war field of today as represented by the network centric warfare(NCW), the reliability of weapon systems is definitely crucial. In this context, it is inevitable to develop software reliably enough to make the weapon systems operate robustly in the combat field. The reliability engineering activities performed to develop software in the domestic area seem to be limited to the software reliability estimations for some projects. To ensure that the target reliability of software be maintained through the system's development period, a more systematic approach to performing software reliability engineering activities are necessary from the beginning of the development period. In this paper, we consider the software reliability in terms of the development of a weapon system as a whole. Thus, from the systems engineering point of view, we analyze the models and methods that are related to software reliability and a variety of associated activities. As a result, a process is developed, which can be called the software reliability engineering process for weapon systems (SREP-WS), The developed SREP-WS can be used in the development of a weapon system to meet a target reliability throughout its life-cycle. Based on the SREP-WS, the software reliability could also be managed quantitatively.