• Journal of the Korean Society for Railway
• /
• v.16 no.4
• /
• pp.331-339
• /
• 2013

This study highlights the theme of human error emerging as a critical issue in the railroad industry, conducting exploratory research on the effects of health, cognition, and safety climates on safety behavior and accidents using an empirical method. The statistical results based on questionnaires received from 204 train drivers indicate that psychological fatigue, cognitive failure, and internal locus of control as individual variables and CEO philosophy and behavior of immediate boss as organizational variables have significant relationships with safety behavior, while cognitive failure, CEO philosophy, behavior of immediate boss, and education were found to be significant variables with respect to accidents. Furthermore, unsafe behavior such as mistakes and violations showed negative effects on near misses and responsibility accidents, respectively. Based on these results, effective alternatives and countermeasures needed to mitigate human error were posited.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.9
• /
• pp.3583-3597
• /
• 2020

In this paper, a parallel deep learning model using a convolutional neural network and a dilated convolutional neural network is proposed to classify Alzheimer's disease with high accuracy in PET/CT images. The developed model consists of two pipelines, a conventional CNN pipeline, and a dilated convolution pipeline. An input image is sent through both pipelines, and at the end of both pipelines, extracted features are concatenated and used for classifying Alzheimer's disease. Complimentary abilities of both networks provide better overall accuracy than single conventional CNNs in the dataset. Moreover, instead of performing binary classification, the proposed model performs three-class classification being Alzheimer's disease, mild cognitive impairment, and normal control. Using the data received from Dong-a University, the model performs classification detecting Alzheimer's disease with an accuracy of up to 95.51%.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.9 no.3
• /
• pp.921-940
• /
• 2015

In this paper, we investigate the multi-resource allocation problem, a unique feature of which is that the multiple resources can compensate each other while achieving the desired system performance. In particular, power and time allocations are jointly optimized with the target of energy efficiency under the resource-limited constraints. Different from previous studies on the power-time tradeoff, we consider a multi-server case where the concurrent serving users are quantitatively restricted. Therefore user selection is investigated accompanying the resource allocation, making the power-time tradeoff occur not only between the users in the same server but also in different servers. The complex multivariate optimization problem can be modeled as a variant of 2-Dimension Bin Packing Problem (V2D-BPP), which is a joint non-linear and integer programming problem. Though we use state decomposition model to transform it into a convex optimization problem, the variables are still coupled. Therefore, we propose an Iterative Dual Optimization (IDO) algorithm to obtain its optimal solution. Simulations show that the joint multi-resource allocation algorithm outperforms two existing non-joint algorithms from the perspective of energy efficiency.

• Journal of Korea Game Society
• /
• v.16 no.2
• /
• pp.27-40
• /
• 2016

Predicting stealthy behaviors plays an important role in designing stealth games. It is, however, difficult to automate this task because human players interact with dynamic environments in real time. In this paper, we present a reinforcement learning (RL) method for simulating stealthy movements in dynamic environments, in which an integrated model of Q-learning with Artificial Neural Networks (ANN) is exploited as an action classifier. Experiment results show that our simulation agent responds sensitively to dynamic situations and thus is useful for game level designer to determine various parameters for game.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.4 no.3
• /
• pp.273-288
• /
• 2010

In this paper, we propose a cooperative multi-relay scheme for a secondary system to achieve spectrum access along with a primary system. In the primary network, a primary transmitter (PT) transmits the primary signal to a primary receiver (PR). In the secondary network, N secondary transmitter-receiver pairs (ST-SR) selected by a centralized control unit (CCU) are ready to assist the primary network. In particular, in the first time slot, PT broadcasts the primary signal to PR, which is also received by STs and SRs. At STs, the primary signal is regenerated and linearly combined with the secondary signal by assigning fractions of the available power to the primary and secondary signals respectively. The combined signal is then broadcasted by STs in a predetermined order. In order to achieve diversity gain, STs, SRs and PT will combine received replicas of the primary signal, using selection combining technique (SC). We derive the exact outage probability for the primary network as well as the secondary network. The simulation results are presented to verify the theoretical analyses.

• Journal of Computing Science and Engineering
• /
• v.7 no.2
• /
• pp.112-121
• /
• 2013

The last years have seen a rise of interest in using electroencephalography-based brain computer interfacing methodology for investigating non-medical questions, beyond the purpose of communication and control. One of these novel applications is to examine how signal quality is being processed neurally, which is of particular interest for industry, besides providing neuroscientific insights. As for most behavioral experiments in the neurosciences, the assessment of a given stimulus by a subject is required. Based on an EEG study on speech quality of phonemes, we will first discuss the information contained in the neural correlate of this judgement. Typically, this is done by analyzing the data along behavioral responses/labels. However, participants in such complex experiments often guess at the threshold of perception. This leads to labels that are only partly correct, and oftentimes random, which is a problematic scenario for using supervised learning. Therefore, we propose a novel supervised-unsupervised learning scheme, which aims to differentiate true labels from random ones in a data-driven way. We show that this approach provides a more crisp view of the brain states that experimenters are looking for, besides discovering additional brain states to which the classical analysis is blind.

• Journal of Communications and Networks
• /
• v.15 no.5
• /
• pp.469-475
• /
• 2013

In this paper, we propose an efficient interference management technique which controls the number of resource blocks (or subcarriers) shared with other cells based on statistical interference levels among cells. The proposed technique tries to maximize average throughput of a femto cell user under a constraint on non-real time control of a femto cell network while guaranteeing a target throughput value of a macro cell user. In our proposed scheme, femto cells opportunistically use resource blocks allocated to other cells if the required average user throughput is not attained with the primarily allocated resource blocks. The proposed method is similar to the underlay approach in cognitive radio systems, but resource block sharing among cells is statistically controlled. For the statistical control, a femto cell sever constructs a table storing average mutual interference among cells and periodically updates the table. This statistical approach fully satisfies the constraint of non-real time control for femto cell networks. Our simulation results show that the proposed scheme achieves higher average femto user throughput than conventional frequency reuse schemes for time varying number of users.

• Proceedings of the ESK Conference
• /
• 1996.10a
• /
• pp.185-190
• /
• 1996

Navigation system is regarded as the interface border line between the Intelligent Transportation Systems (ITS) and the driver as the prospective information provider of the ATIS (Advanced Traveler Information System). Following theory, if the navigation system appropriately designed and utilized, that can maximize the transport efficiency, contribute to improvements of the environments and road safety. To accomplish these dinds of objectives of the navigation system use, human factors plays an important roles specially focused on the driver's safety, performance and system usability. Because the effectiveness of the system depends on the acceptance of the system, and the extent to which the system conforms to driver physical and cognitive limitations and capabilities. Therefore, the ergonomic design vaniables must be seriously selected and reflected in early design step for more effective and appreciate product design. As the first step of this aim, this study selected and categorized the human factors design variables of the navigation system.

• Journal of Information Processing Systems
• /
• v.10 no.3
• /
• pp.443-458
• /
• 2014

An extreme learning machine (ELM) is a recently proposed learning algorithm for a single-layer feed forward neural network. In this paper we studied the ensemble of ELM by using a bagging algorithm for facial expression recognition (FER). Facial expression analysis is widely used in the behavior interpretation of emotions, for cognitive science, and social interactions. This paper presents a method for FER based on the histogram of orientation gradient (HOG) features using an ELM ensemble. First, the HOG features were extracted from the face image by dividing it into a number of small cells. A bagging algorithm was then used to construct many different bags of training data and each of them was trained by using separate ELMs. To recognize the expression of the input face image, HOG features were fed to each trained ELM and the results were combined by using a majority voting scheme. The ELM ensemble using bagging improves the generalized capability of the network significantly. The two available datasets (JAFFE and CK+) of facial expressions were used to evaluate the performance of the proposed classification system. Even the performance of individual ELM was smaller and the ELM ensemble using a bagging algorithm improved the recognition performance significantly.

• Korean Journal of Computational Design and Engineering
• /
• v.14 no.6
• /
• pp.424-433
• /
• 2009

In order to improve productivity and efficiency of ship production process, production technology converged with Information Technology can be considered. Mid-term scheduling based on long-term schedule of ship building and execution planning based on short-term production schedule have an important role in ship production processes and techniques. However, data used in the scheduling are from the experiences of the past, cognitive, and often inaccurate, moreover the updates of the data by formatted documents are not being performed efficiently. This paper designs the tracking and monitoring system for the curved plates forming process with shop level. At first step to it, we redefine and analyze the curved plates forming process by using SysML. From the definition and analysis of the curved plates forming process, we design the system with respect to operational view considering operational environment and interactions between systems included and scenario about operation, and with respect to system view considering functionalities and interfaces of the system. In order to study the feasibility of the system designed, a prototype of the system has been implemented with 13.56 MHz RHD hardware and application software.