• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.64 no.5
• /
• pp.779-785
• /
• 2015

Recently, home energy management system (HEMS) for power consumption reduction has been widely used and studied. The HEMS performs electric power consumption control for the indoor electric device connected to the HEMS. However, a traditional HEMS is used for passive control method using some particular power saving devices. Disadvantages with this traditional HEMS is that these power saving devices should be newly installed to build HEMS environment instead of existing home appliances. Therefore, an HEMS, which performs with existing home appliances, is needed to prevent additional expenses due to the purchase of state-of-the-art devices. In this paper, an intelligent inference algorithm for EMS at home for non-power saving electronic equipment, called legacy devices, is proposed. The algorithm is based on the adaptive network fuzzy inference system (ANFIS) and has a subsystem that notifies retraining schedule to the ANFIS to increase the inference performance. This paper discusses the overview and the architecture of the system, especially in terms of the retraining schedule. In addition, the comparison results show that the proposed algorithm is more accurate than the classic ANFIS-based EMS system.

• 제어로봇시스템학회:학술대회논문집
• /
• 2000.10a
• /
• pp.484-484
• /
• 2000

The problem of determining the proper size of an neural network is recognized to be crucial, especially for its practical implications in such important issues as learning and generalization. Unfortunately, it usually is not obvious what size is best: a system that is too snail will not be able to learn the data while one that is just big enough may learn the slowly and be very sensitive to initial conditions and learning parameters. One popular technique is commonly known as pruning and consists of training a larger than necessary network and then removing unnecessary weights/nodes. In this paper, a new pruning method is developed, based on the penalty-term methods. This method makes the neural network good for the generalization and reduces the retraining time after pruning weights/nodes.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.28 no.7C
• /
• pp.705-711
• /
• 2003

This paper studies the application of a fuzzy-ARTMAP neural network to digital communications channel equalization. This approach provides new solutions for solving the problems, such as complexity and long training, which found when implementing the previously developed neural-basis equalizers. The proposed fuzzy-ARTMAP equalizer is fast and easy to train and includes capabilities not found in other neural network approaches; a small number of parameters, no requirements for the choice of initial weights, automatic increase of hidden units, no risk of getting trapped in local minima, and the capability of adding new data without retraining previously trained data. In simulation studies, binary signals were generated at random in a linear channel with Gaussian noise. The performance of the proposed equalizer is compared with other neural net basis equalizers, specifically MLP and RBF equalizers.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.37 no.3A
• /
• pp.172-178
• /
• 2012

This paper studies the application of a fuzzy-ARTMAP (FAM) neural network to multi-user detector (MUD) for direct sequence (DS)-code division multiple access (CDMA) system. This method shows new solution for solving the problems, such as complexity and long training, which is found when implementing the previously developed neural-basis MUDs. The proposed FAM based MUD is fast and easy to train and includes capabilities not found in other neural network approaches; a small number of parameters, no requirements for the choice of initial weights, automatic increase of hidden units, no risk of getting trapped in local minima, and the capabilities of adding new data without retraining previously trained data. In simulation studies, binary signals were generated at random in a linear channel with Gaussian noise. The performance of FAM based MUD is compared with other neural net based MUDs in terms of the bit error rate.

• Journal of Korean Clinical Health Science
• /
• v.1 no.1
• /
• pp.47-61
• /
• 2013

Purpose. Effective health policy to raise education rate and to provide basic data to identify the college of Education degree and CPR. Purpose of this study was to inform the need for retraining of college students received CPR training. Methods. The sample consisted of 70 a series of health related university students and 70 the general college students ; total 120 in J city. The period of data collecting was from November 1st to Nov. 16th. The tools were 24 questionnaires named "CPR Survey". The collected data were analyzed to get frequency, percentage, average, and standard deviation, t-test and Person's correlation coefficients by using of SPSS for WINDOW 12.0 K program. Results. The number of CPR-trained persons was significantly higher in the health related university students than general college students. Conclusions. CPR training after the passage of time, the percentage of correct answer got lower as time goes by. The answer should be 'School formal education' was the highest. The percentage of students who recognized the necessity of CPR re-education was high.

• Journal of the Korean Society for Library and Information Science
• /
• v.52 no.1
• /
• pp.359-380
• /
• 2018

In the digital society where ICT technology is highly developed, digital literacy is an essential competence for working and living. Developed countries around the world have been working hard to solve the digital divide and improve digital literacy. In this paper, we investigated and analyzed the case of US public libraries for improvement of digital literacy of adults including the older people. To do this, we analyzed the educational program type of digital literacy, education method, and the contents of the program, focusing on the best practices libraries of the program. Many of the educational programs still related to basic computer and Internet technologies, and training programs on Micro Office, e-mail, social media, and smartphone and tablet computing were also high. The most frequent and daily training method was informal point of use, and the content and level of education appeared to be very diverse. For digital literacy training, the librarians of the public library considered librarians' digital competence and retraining to be the most important, and the library facility and the latest equipment to be suitable for the operation of the digital literacy education program.

• Convergence Security Journal
• /
• v.16 no.2
• /
• pp.35-53
• /
• 2016

For the development of national competency standards to successfully perform the duties by systemizing the contents of education and training such as knowledge technology grounding required to perform the duties in the industrial field, the necessity has been raised in that it can solve the time and cost of re-investment through retraining and train high-quality human resources required in the industrial field. For the development of the education and training process with a focus on duty performers for the security sector classifying work and duties into 1:1 among sub-categories based on the NCS classification system, this study enhances the site suitability of security sector education and proposes the development of the practical education and training process required by the industry by using NCS development manual and DACUM technique, the worker-centered job analysis considering 'correlation' with the actual work, 'applicability' of progress and 'efficiency' for costs and time.

• Parasites, Hosts and Diseases
• /
• v.47 no.2
• /
• pp.179-183
• /
• 2009

Prevalence rates reported for malaria in pregnancy in Nigeria vary considerably. The accuracy of results of malaria diagnosis is dependent on training, experience, and motivation of the microscopist as well as the laboratory facility available. Results of training programmes on malaria microscopy have shown low levels of sensitivity and specificity of those involved in malaria diagnosis routinely and for research. This study was done to ascertain the true prevalence of malaria in pregnancy in Lagos, South-West Nigeria. A total of 1,084 pregnant women were recruited into this study. Blood smears stained with Giemsa were used for malaria diagnosis by light microscopy. Malaria infection during pregnancy presents mostly as asymptomatic infection. The prevalence of malaria in this population was 7.7% (95% confidence interval; 6.2-9.4%). Factors identified to increase the risk of malaria infection include young maternal age (<20 years), and gravidity (primigravida). In conclusion, this study exposes the over-diagnosis of malaria in pregnancy and the need for training and retraining of laboratory staffs as well as establishing the malaria diagnosis quality assurance programme to ensure the accuracy of malaria microscopy results at all levels.

• Journal of Engineering Education Research
• /
• v.8 no.2
• /
• pp.16-23
• /
• 2005

In knowledge-based information society, national competitive power is determined by human resources rather than by material resources. Therefore, training technicians with new skills has been gathering as a question Polytechnic colleges play an important role in training technicians as a 2 year courses educational institute conducting both master craftsmen and craftsmen and occupational training. At present, an electricity departments of polytechnic colleges should look at the reality of retraining which is taken more time and more budget for graduates' employment in the industries, as reorganizing in knowledge based school subject they also train the students to be the excellent poly technicians who have both theory and practice by reflecting new technology in developing curriculum needed at the jobsite to join in upgrading our country's competitiveness as a go-between technician at the jobsite. In this paper, a four-semester system was investigated to meet the condition of colleges educating the skilled technicians needed the industry by analyzing the status of polytechnic colleges.

• Journal of Information Processing Systems
• /
• v.19 no.1
• /
• pp.109-117
• /
• 2023

Defect detection is one of the key factors in fabric quality control. To improve the speed and accuracy of denim fabric defect detection, this paper proposes a defect detection algorithm based on cascading feature extraction architecture. Firstly, this paper extracts these weight parameters of the pre-trained VGG16 model on the large dataset ImageNet and uses its portability to train the defect detection classifier and the defect recognition classifier respectively. Secondly, retraining and adjusting partial weight parameters of the convolution layer were retrained and adjusted from of these two training models on the high-definition fabric defect dataset. The last step is merging these two models to get the defect detection algorithm based on cascading architecture. Then there are two comparative experiments between this improved defect detection algorithm and other feature extraction methods, such as VGG16, ResNet-50, and Xception. The results of experiments show that the defect detection accuracy of this defect detection algorithm can reach 94.3% and the speed is also increased by 1-3 percentage points.