• KIPS Transactions on Computer and Communication Systems
• /
• v.11 no.7
• /
• pp.205-216
• /
• 2022

One of the many tools used for distributed deep learning training is Kubeflow, which runs on Kubernetes, a container orchestration tool. TensorFlow jobs can be managed using the existing operator provided by Kubeflow. However, when considering the distributed deep learning training jobs based on the parameter server architecture, the scheduling policy used by the existing operator does not consider the task affinity of the distributed training job and does not provide the ability to dynamically allocate or release resources. This can lead to long job completion time and low resource utilization rate. Therefore, in this paper we proposes a new operator that efficiently schedules distributed deep learning training jobs to minimize the job completion time and increase resource utilization rate. We implemented the new operator by modifying the existing operator and conducted experiments to evaluate its performance. The experiment results showed that our scheduling policy improved the average job completion time reduction rate of up to 84% and average CPU utilization increase rate of up to 92%.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.10
• /
• pp.1509-1516
• /
• 2022

This paper presents a method to generate a model of the code implementation for UML state diagrams. First, it promotes the states of a state machine into objects, and then it structures the behavior model on the mechanism of a state diagram based on State design pattern. Then, it establishes the rules of generating the code implementation, and using the rules, the Java code mode is generated for the implementations of State Diagrams in Java syntax grammar. In addition, Structuring the information of the code model is necessary for generating Java code automatically. The meta information is composed of Meta-Class Model and Meta-Behavior Model, on which we could construct the automatic code generating engine for UML State Diagrams. The implementation model generation method presented in this paper could be used as a stand-alone engine, or included and integrated as a module in the UML tools.

• Journal of The Korean Association of Information Education
• /
• v.25 no.6
• /
• pp.917-925
• /
• 2021

This study aims to specify a constructivism-based instructional method using a modular data analysis tool. The value and meaning of a modular data analysis tool have been examined to be applied in the national curriculum for artificial intelligence education and the process of cultivating problem-solving ability based on computational thinking. The modular data analysis tool visually expresses the cognitive thinking process that forms the schema in equilibrating through assimilation and adjustment. Artificial intelligence education has features that embody abstract knowledge and structure the data analysis module through the represented schema as a BlackBox implemented as an algorithm. Therefore, the value of the modular data analysis tool could be examined because it has the advantage of connecting the conceptual and implicit schema.

• Journal of Software Assessment and Valuation
• /
• v.16 no.2
• /
• pp.1-8
• /
• 2020

The similarity detection to plagiarism or duplication of computer programs requires a different type of analysis methods and tools according to the programming language used in the implementation and the sort of code to be analyzed. In recent years, the similarity appraisal for the object code in the embedded system, which requires a considerable resource along with a more complicated procedure and advanced skill compared to the source code, is increasing. In this study, we described a method for analyzing the similarity of functional units in the assembly language through the conversion of object code using the reverse engineering approach, such as the reverse assembly technique to the object code. The instruction and operand table for comparing the similarity is generated by using the syntax analysis of the code in assembly language, and a tool for detecting the similarity is designed.

• Journal of Software Assessment and Valuation
• /
• v.17 no.2
• /
• pp.101-107
• /
• 2021

Animation production is collaborated by many experts and gives many ideas for new and interesting video production. Interesting video production is a problem directly related to the success of the project, so it can be said that it is better to create an environment that is not burdened with technical aspects in expressing ideas. In the actual keyframe animation production environment, ideas are frequently modified to obtain better results, and techniques that are re-used so that the animation key pose data developed at the early stage of the possible stage can be rewritten without abandoning it, and functions that can temporarily change the center of gravity contribute to the productivity of animation work and greatly help the creator to improve the creative atmosphere. This study analyzes action animations implemented in computer animation software to examine the factors that hinder actual productivity, and derives the technical concepts that can contribute to the improvement of animation production productivity and the necessity of developing related tools.

• Advances in nano research
• /
• v.12 no.2
• /
• pp.185-195
• /
• 2022

Deep learning is another field of artificial intelligence (AI) utilized for computer aided diagnosis (CAD) and image processing in scientific research. Considering numerous mechanical repetitive tasks, reading image slices need time and improper with geographical limits, so the counting of image information is hard due to its strong subjectivity that raise the error ratio in misdiagnosis. Regarding the highest mortality rate of Lung cancer, there is a need for biopsy for determining its class for additional treatment. Deep learning has recently given strong tools in diagnose of lung cancer and making therapeutic regimen. However, identifying the pathological lung cancer's class by CT images in beginning phase because of the absence of powerful AI models and public training data set is difficult. Convolutional Neural Network (CNN) was proposed with its essential function in recognizing the pathological CT images. 472 patients subjected to staging FDG-PET/CT were selected in 2 months prior to surgery or biopsy. CNN was developed and showed the accuracy of 87%, 69%, and 69% in training, validation, and test sets, respectively, for T1-T2 and T3-T4 lung cancer classification. Subsequently, CNN (or deep learning) could improve the CT images' data set, indicating that the application of classifiers is adequate to accomplish better exactness in distinguishing pathological CT images that performs better than few deep learning models, such as ResNet-34, Alex Net, and Dense Net with or without Soft max weights.

• KIPS Transactions on Computer and Communication Systems
• /
• v.11 no.2
• /
• pp.59-66
• /
• 2022

The Joint Coastal Guard System is composed of a maritime surveillance system and a anti-coastal infiltration system, and is a system in which the Navy is mainly responsible for the maritime and the Army is responsible for the coast. We analyzed the operations effectiveness of the joint coastal guard system, in which various weapon systems of the army and navy are operated in a complex way, to the extent to which successful operation is possible against small targets. The operations effectiveness analysis was conducted by defining the operations effectiveness by operation type, configuring the simulation environment using METT-T elements, establishing the assumptions of the simulation scenario, conducting the simulation and analyzing the simulation results by weather condition. The simulation tools used were NORAM and EADSIM. As a result of the operations effectiveness analysis, the joint coastal guard system currently in operation showed a significant difference in operational success depending on the size of the target and weather conditions. This research can be used as useful data for establishing an effective joint coastal guard system and conducting systematic guard operations.

• Communications for Statistical Applications and Methods
• /
• v.30 no.1
• /
• pp.95-107
• /
• 2023

Purpose: The statistical analysis of point processes on linear networks is a recent area of research that studies processes of events happening randomly in space (or space-time) but with locations limited to reside on a linear network. For example, traffic accidents happen at random places that are limited to lying on a network of streets. This paper applies techniques developed for point processes on linear networks and the tools available in the R-package spatstat to estimate the intensity of traffic accidents in Leon County, Florida. Methods: The intensity of accidents on the linear network of streets is estimated using log-linear Poisson models which incorporate cubic basis spline (B-spline) terms which are functions of the x and y coordinates. The splines used equally-spaced knots. Ten different models are fit to the data using a variety of covariates. The models are compared with each other using an analysis of deviance for nested models. Results: We found all covariates contributed significantly to the model. AIC and BIC were used to select 9 as the number of knots. Additionally, covariates have different effects such as increasing the speed limit would decrease traffic accident intensity by 0.9794 but increasing the number of lanes would result in an increase in the intensity of traffic accidents by 1.086. Conclusion: Our analysis shows that if other conditions are held fixed, the number of accidents actually decreases on roads with higher speed limits. The software we currently use allows our models to contain only spatial covariates and does not permit the use of temporal or space-time covariates. We would like to extend our models to include such covariates which would allow us to include weather conditions or the presence of special events (football games or concerts) as covariates.

• KIPS Transactions on Computer and Communication Systems
• /
• v.12 no.6
• /
• pp.197-208
• /
• 2023

In the coming future, anyone using the Internet will have more than one NFT. Unlike FT, NFT can specify the owner, and tracking management is easier than FT. Even in the 2022 survey, WPA2 is the most widely used wireless protocol worldwide to date. As it is a protocol that came out in 2006, it is a protocol with various vulnerabilities at this time. In order to use WPA2-EAP or WPA3 (2018), which were released to compensate for the vulnerabilities of WPA2, additional equipment upgrades are required for STA (station) and AP (access point, router), which are connected devices. The use of expensive router equipment solves the security part, but it is economically inefficient to be introduced in Small Office Home Office (SOHO). This paper uses NFT as a means of authentication and uses the existing WPA2 as it is without equipment upgrade, defend crack tools of WPA2 that have been widely used so far and compared to the existing WPA2, it was shown that it was not difficult to actually use them in SOHO.

• KIPS Transactions on Computer and Communication Systems
• /
• v.12 no.8
• /
• pp.235-242
• /
• 2023

In recent years, as online counseling for infants and adolescents has increased, CNN-based deep learning models are widely used as assistance tools for emotion recognition. However, since most emotion recognition models are trained on mainly adult data, there are performance restrictions to apply the model to infants and adolescents. In this paper, in order to analyze the performance constraints, the characteristics of facial expressions for emotional recognition of infants and adolescents compared to adults are analyzed through LIME method, one of the XAI techniques. In addition, the experiments are performed on the male and female groups to analyze the characteristics of gender-specific facial expressions. As a result, we describe age-specific and gender-specific experimental results based on the data distribution of the pre-training dataset of CNN models and highlight the importance of balanced learning data.