• Journal of Navigation and Port Research
• /
• v.35 no.3
• /
• pp.213-218
• /
• 2011

This study simulates and compares the evacuation time and characteristics according to the living patterns on board a training ship which was launched in Dec. 2005, on the viewpoint of the various activities being possible on board a cruise ship. Based on interviews with personnels on board, 3 living patterns are set as representative living conditions; Pattern A(all personnels are positioned at their cabins), Pattern B(all personnels are positioned at lecture rooms, offices or else), Pattern C(all personnels are positioned at restaurant or cafeteria). The simulation results show that Pattern B is comparatively ideal because the evacuation time is short and there is less delay of personnels' movement on each deck. On the contrary, Pattern C is evaluated as the worst because the average evacuation time took more than 360 seconds and the bottle-neck happened at Upper deck. As a result, this study proposes the needs of various countermeasures against the fire and/or disaster, considering the various living patterns on cruiser(s) and/or passenger ship(s).

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.12 no.3
• /
• pp.201-207
• /
• 2012

A RAM-based neural network is a weightless neural network based on binary neural network(BNN) which is efficient neural network with a one-shot learning. RAM-based neural network has multiful information bits and store counts of training in BNN. Supervised learning based on the RAM-based neural network has the excellent performance in pattern recognition but in pattern categorization with unsupervised learning as unsuitable. In this paper, we propose a unsupervised learning algorithm in the RAM-based neural network to perform pattern categorization. By the proposed unsupervised learning algorithm, RAM-based neural network create categories depending on the input pattern by itself. Therefore, RAM-based neural network for supervised learning and unsupervised learning should proof of all possible complex models. The training data for experiments provided by the MNIST offline handwritten digits which is consist of 0 to 9 multi-pattern.

• The Journal of Korean Physical Therapy
• /
• v.26 no.4
• /
• pp.249-256
• /
• 2014

Purpose: The aim of this study was to assess the effects of proprioceptive neuromuscular facilitation (PNF) pattern exercise using sprinter and skater on balance and gait in the stroke patients. Methods: Twenty-two subjects were randomly assigned to the experimental group (n=11) and the control group (n=11). The experimental group performed PNF pattern exercise using sprinter and skater for 15 minutes with conventional physical therapy for 35 minutes (matt and gait training for 15 minutes + FES stimulation for 20 minutes), while the control group performed only conventional physical therapy for 50 minutes (matt and gait training for 30 minutes + FES stimulation for 20 minutes). Both groups performed therapeutic interventions for five days per week, for a period of four weeks. Functional Reach Test (FRT) and Berg Balance Scale (BBS) were used for assessment of balance, and Timed-Up and Go test (TUG) was used for testing of gait. Results: The experimental group showed significant improvements in the FRT and the BBS, while the control group did not show significant changes in two measurements. The experimental group also showed significant improvements, however, the control group did not show significant changes in the TUG. In post-values of three measurements, significant differences were observed between the two groups (p<0.05). Conclusion: This study demonstrated that PNF pattern exercise using sprinter and skater may be used to improve balance and gait function in stroke patients. Thus, we suggested that PNF pattern exercise using sprinter and skater would be a therapeutic intervention in stroke rehabilitation.

• PNF and Movement
• /
• v.13 no.3
• /
• pp.127-134
• /
• 2015

Purpose: The current study seeks to examine the effect of coordinative pattern exercises of the upper and lower extremities using harnesses and walking rails on the walking and balance abilities of chronic stroke patients, and to develop effective programs and training methods to improve the functions of such patients. Methods: Subjects included 16 patients with hemiplegia caused by stroke. The subjects were randomly divided into an experimental group (n=8), on which coordinative pattern exercises of the upper and lower extremities were conducted, and a comparison group (n=8) that received typical exercise therapy. The experimental group underwent 30 minutes of typical exercise therapy and 30 minutes of coordinative pattern exercises of the upper and lower extremities, while the comparison group underwent typical exercise therapy for 30 minutes twice a day, five days per week for a six-week period. To evaluate walking ability, 10-m walking tests (10MWT) and 6-m walking tests (6MWT) were conducted. To assess balance ability, timed up and go tests (TUG) were performed. Results: After the intervention, significant (p<0.05) differences were seen in the 10MWT, 6MWT, and TUG in both the experimental and comparison groups. As for the 10MWT, the experimental group showed more significant improvement than the comparison group (p<0.05). In terms of the 6MWT, no significant differences were found between the groups, while the experimental group showed more significant differences than the comparison group in the TUG (p<0.05). Conclusion: The results from the current research indicate that training programs that apply coordinative pattern exercises of the upper and lower extremities with harnesses are extremely effective for improving the walking and balance abilities of chronic stroke patients.

• Journal of rehabilitation welfare engineering & assistive technology
• /
• v.10 no.4
• /
• pp.329-336
• /
• 2016

Deaf people using sign language are experiencing social inequalities and financial losses due to communication restrictions. In this paper, real-time pattern recognition algorithm was applied to distinguish American Sign Language using an armband sensor(8-channel EMG sensors and one IMU) to enable communication between the deaf and the hearing people. The validation test was carried out with 11 people. Learning pattern classifier was established by gradually increasing the number of training database. Results showed that the recognition accuracy was over 97% with 20 training samples and over 99% with 30 training samples. The present study shows that sign language recognition using armband sensor is more convenient and well-performed.

• International Journal of Industrial Entomology and Biomaterials
• /
• v.3 no.2
• /
• pp.187-190
• /
• 2001

Inheritance pattern of resistance to Bombyx mori nuclear polyhedrosis virus (BmNPV) was studied in an Indian silkworm stock TX by single back-cross test method. The resistant parent ［TX］, susceptible parent ［HM］, their Fl, F2, and Fl progeny back-crossed to TX [BC(R)] and HM [BC(S)] were inoculated per os with a fixed concentration of BmNPV($0.5{\times}10^{th} PIB/ml$) on the first day of second stadium. The cumulative mortality was recorded until day $10^{\times}$ post-inoculation. The results show that the resistance to BmNPV in TX fellow mono Mendelian inheritance pattern. The resistance dominated over the susceptibility at Fl. At F2, the resistant and susceptible offspring segregated in 3:1 ratio whereas at BC(S), the resistant and susceptible offspring segregated in 1:1 ratio. The response of BC(R) was more or less like the resistant parent TX which confirms the involvement of a major dominant gene conferring resistance to BmNPV in TX. The possible mechanism of inheritance of resistance in TX is discussed.

• The Journal of Korean Institute of Information Technology
• /
• v.17 no.12
• /
• pp.29-38
• /
• 2019

A path prediction method using lifelog requires a large amount of training data for accurate path prediction, and the path prediction performance is degraded when the training data is insufficient. The lack of training data can be solved using data of other users having similar user movement patterns. Therefore, this paper proposes a path prediction algorithm based on user similarity. The proposed algorithm learns the path in a triple grid pattern and measures the similarity between users using the cosine similarity technique. Then, it predicts the path with applying measured similarity to the learned model. For the evaluation, we measure and compare the path prediction accuracy of proposed method with the existing algorithms. As a result, the proposed method has 66.6% accuracy, and it is evaluated that its accuracy is 1.8% higher than other methods.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.31B no.4
• /
• pp.145-154
• /
• 1994

This paper proposes an efficient method for improving the training performance of the neural network by using a hybrid of a stochastic approximation and a backpropagation algorithm. The proposed method improves the performance of the training by appliying a global optimization method which is a hybrid of a stochastic approximation and a backpropagation algorithm. The approximate initial point for a stochastic approximation and a backpropagation algorihtm. The approximate initial point for fast global optimization is estimated first by applying the stochastic approximation, and then the backpropagation algorithm, which is the fast gradient descent method, is applied for a high speed global optimization. And further speed-up of training is made possible by adjusting the training parameters of each of the output and the hidden layer adaptively to the standard deviation of the neuron output of each layer. The proposed method has been applied to the parity checking and the pattern classification, and the simulation results show that the performance of the proposed method is superior to that of the backpropagation, the Baba's MROM, and the Sun's method with randomized initial point settings. The results of adaptive adjusting of the training parameters show that the proposed method further improves the convergence speed about 20% in training.

• Journal of KIISE:Software and Applications
• /
• v.27 no.2
• /
• pp.97-107
• /
• 2000

Incremental learning generally reduces training time and increases the generalization of a neural network by selecting training data incrementally during the training. However, the existing methods of incremental learning repeatedly evaluate the importance of training data every time they select additional data. In this paper, an incremental learning algorithm is proposed for pattern classification problems. It evaluates the importance of each piece of data only once before starting the training. The importance of the data depends on how close they are to the decision boundary. The current paper presents an algorithm which orders the data according to their distance to the decision boundary by using clustering. Experimental results of two artificial and real world classification problems show that this proposed incremental learning method significantly reduces the size of the training set without decreasing generalization performance.

• The Journal of Korean Medicine
• /
• v.42 no.4
• /
• pp.40-60
• /
• 2021

Objectives: This study aimed to evaluate the impact of a training course on pharmacovigilance for future doctors of Korean medicine (DKM). Methods: In 2020, a pharmacovigilance training course was conducted for 57 senior students of a Korean medicine (KM) college, and its impact to the students were assessed in terms of the knowledge, attitudes, and perceptions of these students at three-time points: pre-training, post-training, and 4-6 months after the end of the training. Results: A total of 38 students completed the survey. The average score from the knowledge test increased significantly after training compared to prior to the training (5.47±2.140, 6.61±1.001, respectively, p<0.001) and was maintained until the final survey (6.61±1.220). The rate of correct answers to most of the knowledge test questions increased after the training but decreased in the final survey. In terms of attitudes, self-confidence in causality assessment (2.63±1.025, 4.58±0.826, p<0.001) and spontaneous reporting (2.08±1.050, 4.74±0.446, p<0.001) significantly increased after the training and then slightly decreased (3.92±1.171, 4.40±0.755). The perception level was high prior to the training, and this pattern was maintained throughout the study period. Students responded that pharmacovigilance education was necessary for DKM after training, and for the undergraduates of KM colleges. Conclusions: This study shows that this pharmacovigilance training course is effective for students majoring in KM but that retraining is required at least 6 months after the initial training. Further follow-up studies are needed to ensure that students actively participate in spontaneous reporting after graduation, and continuous education should be provided to graduates.