• KIPS Transactions on Software and Data Engineering
• /
• v.10 no.9
• /
• pp.375-384
• /
• 2021

Deep learning techniques have been proven to have high performance in image processing and are applied in various fields. The most widely used methods for validating a deep learning model include a holdout verification method, a k-fold cross verification method, and a bootstrap method. These legacy methods consider the balance of the ratio between classes in the process of dividing the data set, but do not consider the ratio of various features that exist within the same class. If these features are not considered, verification results may be biased toward some features. Therefore, we propose a deep learning model validation method based on data feature coverage for image classification by improving the legacy methods. The proposed technique proposes a data feature coverage that can be measured numerically how much the training data set for training and validation of the deep learning model and the evaluation data set reflects the features of the entire data set. In this method, the data set can be divided by ensuring coverage to include all features of the entire data set, and the evaluation result of the model can be analyzed in units of feature clusters. As a result, by providing feature cluster information for the evaluation result of the trained model, feature information of data that affects the trained model can be provided.

• Proceedings of the Korean Society of Machine Tool Engineers Conference
• /
• 2000.04a
• /
• pp.58-65
• /
• 2000

This study is concerned about the verification and the implementation of a mechanical model for the force system in end milling. The model is based on the relationship between the cutting forces and the chip thickness. The components of the model are based on the average cutting forces which are experimentally obtained. And, both instantaneous and average force system characteristics are described as a function of cut geometry and a feed rate. This model employed two specific cutting forces, instantaneous and average specific cutting force, and the models which obtained using two cutting forces were compared and analyzed. In this study, cutter deflection with respect to the center of rotation is considered, which is a major part of the tool run-outs. The effect of run-out on the cutting forces is also discussed. The relationships among the run-out parameters, cutting parameters and the resulting force system characteristics are presented. In all cases, for the down milling with a right hand helix cutter is considered.

• Journal of The Korean Society of Agricultural Engineers
• /
• v.46 no.7
• /
• pp.23-32
• /
• 2004

A spatial simulation model for rural and urban sprawl phenomena was developed with GIS and cellular automata techniques. The model finds out built-up areas invading toward rural areas required for development of existing urban area. Probability of land use change for optimizing the development area was determined using a land suitability analysis method interfaced with GIS methods, based on several criteria in terms of geographic and accessibility factors such as slope of land and distance from city center. Weighting values of the criteria were quantified by an analytic hierarchy process method. For model applicability test, the parameters of criteria were calibrated based on the changes in time series land use data of the test city for 1986, 1996, and 2000, which were classified by remote sensing techniques. Simulated and observed areas in land use maps for city shape of 1996 showed good similarities with each other through a morphology verification method. The model enabled us to evaluate the spatial expansion phenomena of cities considering boundary conditions, and also to simulate land use planning for rural areas in urban fringe.

• Journal of Environmental Impact Assessment
• /
• v.18 no.5
• /
• pp.271-280
• /
• 2009

The objective of this study is to develop a reliable flow forecasting model based on neural network algorithm in order to provide flow rate at stream sections without flow measurement in Nakdong river. Stream flow rate measured at 8-days interval by Nakdong river environment research center, daily upper dam discharge and precipitation data connecting upstream stage gauge were used in this development. Back propagation neural network and multi-layer with hidden layer that exists between input and output layer are used in model learning and constructing, respectively. Model calibration and verification is conducted based on observed data from 3 station in Nakdong river.

• Proceedings of the KAIS Fall Conference
• /
• 2003.11a
• /
• pp.50-56
• /
• 2003

In the development of complex software system, it is important to use hierarchical use case model due to the complex scope of development procedure. The use case model is core factor of the OMG (Object Management Group)'s UML (Unified Modeling Language) diagrams. In this paper, we propose a novel method to check syntactic consistency automatically in use case models at the different level of abstraction. This method is a rule-based approach which utilizes actor tree, use case tree and use case description. The proposed method is simulated on ITS (Intelligent Transportation System) architecture for the verification.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.4
• /
• pp.1232-1245
• /
• 2021

In this study, prediction of product sales as they relate to changes in temperature is proposed. This model uses long short-term memory (LSTM), which has shown excellent performance for time series predictions. For verification of the proposed sales prediction model, the sales of short pants, flip-flop sandals, and winter outerwear are predicted based on changes in temperature and time series sales data for clothing products collected from 2015 to 2019 (a total of 1,865 days). The sales predictions using the proposed model show increases in the sale of shorts and flip-flops as the temperature rises (a pattern similar to actual sales), while the sale of winter outerwear increases as the temperature decreases.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.37 no.4
• /
• pp.359-366
• /
• 2009

MATLAB Autocode generation is a feature that converts a block diagram model in Simulink to a c program. Utilizing this function makes MATLAB/Simulink an integrated developing environment, from controller design to implementation. It can reduce development cost and time significantly. However, this automated process requires high reliability on the software, especially the original Simulink block diagram model. And thus, the verification of the codes becomes important. In this study, a UAV flight program which is generated with Simulink is validated and modified according to DO-178B. As a result of applying the procedures, the final program not only satisfied the functional requirement but is also verified with structural point of view with Decision Coverage 93%, Condition Coverage 95% and MC/DC 90%.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.48 no.3
• /
• pp.76-81
• /
• 2011

Intrusion prevention systems (IPS) are important solution about solved problems for inside system security or outsider attacks. When introduce this system, first consideration item is secured rather than multiple function. Colored Petri Nets (CPNs) used that in order to secured verification for user authentication function of intrusion prevention system security model. CPNs is a graphical modeling language suitable for modeling distributed, concurrent, deterministic or non-deterministic systems with synchronous. Like these CPNs was expressed every possible state and occurrence graph. Secured of IPS security model was verified because expression every state using CPN tool and as a result of analyzing the occurrence graph was without a loop or interruption.

• Journal of Navigation and Port Research
• /
• v.45 no.5
• /
• pp.231-237
• /
• 2021

Enhancing the performance of maritime wireless communication has been highlighted by the issue of cell planning in the sea area because of lack of an appropriate Propagation Loss Model (PLM). To resolve the cell planning issue in vast sea areas, it was essential to develop the (PLM) matching the intended sea area. However, there were considerable gaps between the prediction of legacy PLMs and field measurement in propagation loss and there was a need to develop the adjusted PLM (A-PLM). Therefore, cell planning was performed on this adjusted model, including modification of the base station's location, altitude, and antenna azimuth to meet the quality objectives. Furthermore, in order to verify the availability of the cell planning, Communication Service Quality Monitoring System (CS-QMS) was developed in the LTE-Maritime project to collect LTE signal quality information from the onboard equipment at regular intervals and to ensure that the service quality was high enough to satisfy the goals in each designated grid. As a result of verification, the success rate of RSRP was 95.7% for the intensive management zone (IMZ) and 96.4% for the interested zone (IZ), respectively.

• The Korean Journal of Applied Statistics
• /
• v.37 no.2
• /
• pp.209-224
• /
• 2024

Speaker verification is becoming popular as a method of non-face-to-face identity authentication. It involves determining whether two voice data belong to the same speaker. In cases where the criminal's voice remains at the crime scene, it is vital to establish a speaker verification system that can accurately compare the two voice evidence. In this study, to achieve this, a new speaker verification system was built using a deep learning model for Korean language. High-dimensional voice data with a high variability like background noise made it necessary to use deep learning-based methods for speaker matching. To construct the matching algorithm, the ECAPA-TDNN model, known as the most famous deep learning system for speaker verification, was selected. A large dataset of the voice data, Voxceleb, collected from people of various nationalities without Korean. To study the appropriate form of datasets necessary for learning the Korean language, experiments were carried out to find out how Korean voice data affects the matching performance. The results showed that when comparing models learned only with Voxceleb and models learned with datasets combining Voxceleb and Korean datasets to maximize language and speaker diversity, the performance of learning data, including Korean, is improved for all test sets.