• Journal of KIISE:Software and Applications
• /
• v.31 no.5
• /
• pp.537-542
• /
• 2004

Multi-Layer Perceptron network has been mainly applied to many practical problems because of its nonlinear mapping ability. However the generalization ability of MLP networks may be affected by the number of hidden nodes, the initial values of weights and the training errors. These factors, if improperly chosen, may result in poor generalization ability of MLP networks. It is important to identify these factors and their interaction in order to control effectively the generalization ability of MLP networks. In this paper, we have empirically identified the factors that affect the generalization ability of MLP networks, and compared their relative effects on the generalization performance for the conventional and visualized weight selecting methods using the controller box.

• Journal of the Korean Institute of Telematics and Electronics C
• /
• v.36C no.4
• /
• pp.78-86
• /
• 1999

In this paper, a distributed arithmetic digital neural network with learning and testing phase implemented in a body has been studied. The proposed technique is based on the two facts; one is that the weighting coefficients adjusted will be stored in registers without shift, because input values or input patterns are not changed while learning and the other is that the input patterns stored in registers are not changed while testing. The proposed digital neural network is simulated by hardware description language such as VHDL and verified the performance that the neural network was applied to the recognition of seven-segment. To verify proposed neural networks, we compared the learning process of modified perceptron learning algorithm simulated by software with VHDL for 7-segment number recognizer. The results are as follows: There was a little difference in learning time and iteration numbers according to the input pattern, but generally the iteration numbers are 1000 to 10000 and the learning time is 4 to 200$\mu\textrm{s}$. So we knew that the operation of the neural network is learned in the same way with the learning of software simulation, and the proposed neural networks are properly operated. And also the implemented neural network can be built with less amounts of components compared with board system neural network.

• 대한공업교육학회지
• /
• v.37 no.2
• /
• pp.1-19
• /
• 2012

In this study, to figure out the demands for learning, a survey was conducted on learning organization agents participating in learning organization project supported by public fund, through measuring their awareness level of the role and tasks drawn from the precedent study. This questionnaire was developed with four roles and twenty competencies drawn from the precedent study. After conducting the survey on over 200 small and medium corporations participated in learning organization project, the data was analyzed about the role and competency from 123 answer sheets collected. The results were as following First, the awareness of role importance was higher(4.36) than present level(3.73), and was important in order of innovator, operator, spreader, promoter. Second, in competitive importance of learning organization agents, the role of spreader was high, that of promoter was low. Third, in awareness level of learning organization agents, the demand level was generally high(4.1), but acquired level was generally low, so there was a wide gap between the demand level and acquired level. Forth, From the analysis result of those data about work competency of learning organization agents, learning organization promoter and learning leader, KMS operator ability and outcome were come out to be necessary, and next administrative ability about organization changes were come out to be necessary.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.21 no.9
• /
• pp.1744-1751
• /
• 2017

We are striving to strengthen environmental regulations and reduce household waste in all countries around the world. Korea is also striving for the circulation of energy resources by enacting laws to promote resource saving and recycling. The government has implemented an empty bottle deposit system for the recycling of empty bottles, but there is a limit to the collection through manpower and the reverse vending machine is not localized. In this paper, we propose a recyclable bottle recognition and classification algorithm which is essential in the reverser vending machine to promote energy resource circulation. The proposed algorithm is a complex identification algorithm using OpenCV and CNN(Convolution Neural Network). In order to evaluate the effectiveness of the proposed algorithm, we implement a classification system that operates in an reverse vending machine, so that it can easily acquire information about bottles and reverse vending machine in various devices.

• Journal of The Korean Association For Science Education
• /
• v.31 no.5
• /
• pp.758-769
• /
• 2011

This study was designed to investigate the relations of motivation to related variables, and the motivation by student's variables basically through measure of motivation system about science learning. For this, 496 fourth-grade students and 425 eighth-grade students were required to accomplish a questionnaire on behavioral inhibition/activation system about science learning (SL-BIS/BAS), and self-efficacy. Based on these materials, SLBIS/BAS by student's variables and relations between the system to self-efficacy were analyzed. The result of this study shows the following:First, the sensitivity about SL-BIS in the eighth-grade was higher than in the fourth-grade. But sensitivity about SLBAS and self-efficacy in the eighth-grade was lower than in the fourth-grade Hence, the difference in sensitivity about SL-BIS/BAS agree with difference in motivation. Second, there were low negative correlation with SL-BIS and self-efficacy, and positive correlation with SL-BAS and self-efficacy. In addition, two systems account for 41% of self-efficacy, it showed that SL-BIS/BAS relates to self-efficacy. Through the results of this study, relations of motivation to related variables, and the motivation by student's variables were identified. These results would be helpful for teachers to understand the difference about motivation by students' variables and to make a plan for the appropriate strategies for learners.

• Journal of Digital Convergence
• /
• v.18 no.3
• /
• pp.99-110
• /
• 2020

This study deals with the relationship between entrepreneurship of corporate member and creativity at the organizational level by empirically analyzing the relationship between entrepreneurship, organizational learning orientation, individual creativity, and group creativity. In particular, the relationship between entrepreneurship and creativity focuses on analyzing the mediating effects of organizational learning orientation. Based on the literature research, the research model and hypothesis were established by explaining the relationship between entrepreneurship, individual and group creativity, and organizational learning orientation. 308 copies of the questionnaire were distributed and collected for manufacturing workers in Gyeongbuk, and empirical analysis was conducted through structural equations. As a result, it was confirmed that entrepreneurship has an influence on organizational learning orientation without directly affecting individual and group creativity. In addition, the effects of entrepreneurship on organizational learning orientation and organizational learning orientation on individual and group creativity were examined. These findings can provide Directions for organizational management from the cultural perspective by identifying the effects of entrepreneurship and organizational learning orientation at the organizational level.

• Asia-pacific Journal of Multimedia Services Convergent with Art, Humanities, and Sociology
• /
• v.7 no.4
• /
• pp.551-559
• /
• 2017

In this paper, a real-time training method to improve the performance of bearing fault diagnosis. The traditional bearing fault diagnosis cannot classify a condition which is not trained by the classifier. The proposed 4-step method trains and recognizes new condition in real-time, thereby it can classify the condition accurately. In the first step, we calculate the maximum distance value for each class by calculating a Euclidean distance between a feature vector of each class and a centroid of the corresponding class in the training information. In the second step, we calculate a Euclidean distance between a feature vector of new acquired data and a centroid of each class, and then compare with the allowed maximum distance of each class. In the third step, if the distance between a feature vector of new acquired data and a centroid of each class is larger than the allowed maximum distance of each class, we define that it is data of new condition and increase count of new condition. In the last step, if the count of new condition is over 10, newly acquired 10 data are assigned as a new class and then conduct re-training the classifier. To verify the performance of the proposed method, bearing fault data from a rotating machine was utilized.

• Journal of Digital Convergence
• /
• v.10 no.4
• /
• pp.159-165
• /
• 2012

In this paper, the qualification process to target trainees of 200 questionnaires collected 158 call was using the SPSS 18.0 statistical program was statistical results are as follows. First, the demographic characteristics of 158 people a total of 146 people men (92.4%) were 30 and 40 adults. Second, according to gender for women than men showed high satisfaction of sayibeogang. More than a college education group showed a high level of satisfaction. Depending on the marital satisfaction of married people was higher sayibeogang. In addition, cyber-lecture on satisfaction surveys women, highly educated, married, educated in the group were satisfied. 74.5% of total variance, variance, with 88.7 percent reliability was very high reliability. Thus, students' demographic variables, motivation, learning to fit in the field of instructional design and content to support the various efforts will be.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.20 no.12
• /
• pp.83-93
• /
• 2019

The purpose of this study is to suggest implications for maximizing the effectiveness and efficiency of IPP work-learning by studying the effects of the personal characteristics of trainees and firm's field teacher on trainees' employability. The following conclusions were drawn by verifying the causal relationship between factors based on a questionnaire (157 copies) distributed to 200 trainees (learning workers) of 9 universities participating in IPP work-learning training. First, in order to increase the employability of IPP trainees, it is necessary to select trainees with strong core competencies and failure tolerance by verifying individual characteristics of trainees. Secondly, companies should appoint senior staff with strong leadership and teaching capacity as corporate field teachers, and corporate field teacher training institutions should reinforce the training to improve their leadership. Third, work-learning companies should be selected as those with high social integration and absorption capacity, and those companies should try to build an organizational culture of acquiring and learning new technologies. The results of this study are expected to contribute to enhancing the effectiveness of work-learning training by establishing a training system that can systematically select trainees and participating companies and enhance the training capacity of companies and field teachers.

• The Journal of the Korea Contents Association
• /
• v.19 no.2
• /
• pp.66-78
• /
• 2019

This study aimed to investigate the effects of scaffolding on creativity in Wiki-based collaborative learning. Wiki-based collaborative learning was implemented over three sessions among 67 sixth graders in an elementary school, and subjects were divided into three groups: Group 1 had a teachers' scaffolding and self-questioning script; Group 2 had a teachers' scaffolding; and Group 3, the control group, had no scaffolding. Results showed a significant difference in creativity among the groups with different types of scaffolding(Wilks' Lambda=.238, F=8.678, p < .001). Group 1, had significantly higher performance compared to the Group in creativity. However, when self-questoning script and a teachers' scaffolding were offered, self-scaffolding was not found to have a significant effect on learners' Originality of creativity. Group 2 showed higher performance in Originality of creativity when only a teachers' scaffolding was offered in the collaborative learning. The results of this study suggest that teachers' scaffolding and self-scaffolding have positive effects on creativity, but the need for a differentiated self-scaffolding strategy to identify the factors that influence creativity in wiki-based collaborative learning.