• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.47 no.2
• /
• pp.155-167
• /
• 2024

Distribution and logistics industries contribute some of the biggest GDP(gross domestic product) in South Korea and the number of related companies are quarter of the total number of industries in the country. The number of retail tech companies are quickly increased due to the acceleration of the online and untact shopping trend. Furthermore, major distribution and logistics companies try to achieve integrated data management with the fulfillment process. In contrast, small and medium distribution companies still lack of the capacity and ability to develop digital innovation and smartization. Therefore, in this paper, a deep learning-based demand forecasting & recommendation model is proposed to improve business competitiveness. The proposed model is developed based on real sales transaction data to predict future demand for each product. The proposed model consists of six deep learning models, which are MLP(multi-layers perception), CNN(convolution neural network), RNN(recurrent neural network), LSTM(long short term memory), Conv1D-BiLSTM(convolution-long short term memory) for demand forecasting and collaborative filtering for the recommendation. Each model provides the best prediction result for each product and recommendation model can recommend best sales product among companies own sales list as well as competitor's item list. The proposed demand forecasting model is expected to improve the competitiveness of the small and medium-sized distribution and logistics industry.

• Human Ecology Research
• /
• v.62 no.3
• /
• pp.573-583
• /
• 2024

Smart devices have become an essential part of human life; however, excessive dependency on these devices has become a serious issue for both children and adults. In this study, a research model was developed which hypothesized that mothers' smartphone addiction influences children's media dependency, and that this effect is sequentially mediated by mothers' work-family strains and parental monitoring. To test this hypothesis, an analysis of data from the 13th Wave of Panel Study on Korean Children (PSKC), conducted in 2020, was performed. The data were collected from 726 children (359 boys, 367 girls) with an average age of 12.23 years (SD=.30). The analysis was conducted using SPSS 27.0 and Process MACRO Ver. 4.2. The results were as follows. Maternal smartphone addiction increased mothers' work-family strains, which in turn lowered the level of parental monitoring, which was associated with an increase in children's media dependency. Furthermore, mothers' smartphone addiction had a partial mediation effect on children's media dependency. In sum, this study revealed that the mothers' smartphone addiction increased children's media dependency, and this effect was sequentially mediated by an increase in mothers' work-family strains and a decrease in parental monitoring. This study verifies that mothers' smart device usage as well as their working environment has a significant effect on children's media dependency. Therefore, to increase parental monitoring of children's media usage, mothers' work-family strains need to be reduced.

• Journal of Korea Technology Innovation Society
• /
• v.16 no.1
• /
• pp.323-345
• /
• 2013

In this paper, problems with the Korean system regulating the use of university royalties are identified and investigated in order to suggest measures to improve the system in a way that provides a better R&D environment at universities. The Delphi technique was used to gather data from royalty specialists at universities and government ministries. The first Delphi survey conducted used open questions to identify problems in the use of university royalties. Then, closed questions were used for the second Delphi survey. The number of responses and the frequency of answers were analyzed after the first survey, and validity, stability, and reliability analyses were conducted for the second survey. The measures suggested to improve the system regulating the use of university royalties are as follows: First, bonuses for researchers, which are currently 50% or more of collected royalties, need to be decreased, as they are rather high compared to similar bonuses in developed countries, which are around 30% of collected royalties. The guideline for limiting the bonuses, which is explained as XX% or less of collected royalties, is suggested to prevent the excessive use of royalties. Second, rewards for those who contribute to technology transfer and commercialization should be increased. It is also important to build a consensus around the need to reward those who contribute to technology transfer and commercialization. Third, the scale of re-investment into R&D needs to increase. Regulations on royalties should be meaningfully applied to create a positive feedback structure for R&D, which can be described as the process of research, R&D outcomes, technology transfer, collecting royalties, rewarding researchers, and re-investing in R&D. To build a university's R&D capability, re-investment into R&D needs to be regularized as XX% or more of royalties. Fourth, regulations on the royalties of ministries and universities need to be unified. Each category for using royalties needs to be regularized, with detailed matters such as the guideline, process and method for using royalties specified. Also, universities need to make their own specific regulations. Fifth, specific priorities on the use of royalties need to be suggested. Regulation is necessary for the categories that do not have guideline and priorities for the use of royalties. It is hoped that the findings of this research will contribute to reinforcing the R&D capability of universities.

• Journal of Korea Technology Innovation Society
• /
• v.20 no.2
• /
• pp.334-367
• /
• 2017

The ICT industry has been the main driver of Korea's economy with international competitiveness and is expected to be the growth engine that will revitalize the currently depressed economy. A broad range of different perspectives and opinions on the industry exist in Korea and overseas. Some of these are stereotypes, not all of which are based on objective evidence. Stereotypes refer to widely-held fixed opinions on a specific group and do not necessarily have negative connotations. However, they should not be viewed lightly because they can substantially affect decision-making process. In this regard, this study sought to review the stereotypes of ICT industry and identify objective and relative stereotypes. In the study, a decision-tree analysis was conducted on a survey result of 3,300 small and medium-sized enterprises (SMEs) in order to identify Korean ICT companies' characteristics that distinguish them from other technology companies. The decision-tree analysis, a data mining process based on machine learning, took a total of 291 variables into account in 10 subjects such as: corporate business in general, technology development activities as well as organization and people in technology development. Identifying the variables that distinguish ICT companies from other technology companies with the decision-tree analysis, the study then came up with a list of objective stereotypes of ICT companies. The findings from the stereotypes of Korean ICT companies are as follows. First, the companies are in need of technology policies that help R&D planning and market penetration. Second, policies must better support the companies working to sell new products or explore new business. Third, the companies need policies that support secure protection of development outcomes and proper management of IP rights. Fourth, the administrative procedures related to governmental support for ICT companies' R&D projects must be simplified. It is hoped that the outcome of this study will provide meaningful guidance in establishment, implementation and evaluation of technology policies for ICT SMEs, particularly to policymakers or researchers in relevant government agencies who determine R&D policies for ICT SMEs.

• Journal of Intelligence and Information Systems
• /
• v.28 no.1
• /
• pp.263-286
• /
• 2022

This study acquires NBA statistical information for a total of 32 years from 1990 to 2022 using web crawling, observes variables of interest through exploratory data analysis, and generates related derived variables. Unused variables were removed through a purification process on the input data, and correlation analysis, t-test, and ANOVA were performed on the remaining variables. For the variable of interest, the difference in the mean between the groups that advanced to the playoffs and did not advance to the playoffs was tested, and then to compensate for this, the average difference between the three groups (higher/middle/lower) based on ranking was reconfirmed. Of the input data, only this year's season data was used as a test set, and 5-fold cross-validation was performed by dividing the training set and the validation set for model training. The overfitting problem was solved by comparing the cross-validation result and the final analysis result using the test set to confirm that there was no difference in the performance matrix. Because the quality level of the raw data is high and the statistical assumptions are satisfied, most of the models showed good results despite the small data set. This study not only predicts NBA game results or classifies whether or not to advance to the playoffs using machine learning, but also examines whether the variables of interest are included in the major variables with high importance by understanding the importance of input attribute. Through the visualization of SHAP value, it was possible to overcome the limitation that could not be interpreted only with the result of feature importance, and to compensate for the lack of consistency in the importance calculation in the process of entering/removing variables. It was found that a number of variables related to three points and errors classified as subjects of interest in this study were included in the major variables affecting advancing to the playoffs in the NBA. Although this study is similar in that it includes topics such as match results, playoffs, and championship predictions, which have been dealt with in the existing sports data analysis field, and comparatively analyzed several machine learning models for analysis, there is a difference in that the interest features are set in advance and statistically verified, so that it is compared with the machine learning analysis result. Also, it was differentiated from existing studies by presenting explanatory visualization results using SHAP, one of the XAI models.

• Korea Trade Review
• /
• v.42 no.5
• /
• pp.137-159
• /
• 2017

This study divides manufacturing in 18 countries including Korea, China, Japan and OECD countries into 11 areas and estimates and compares the technological efficiency of each industry. The traditional view of productivity is to increase production capacity through technological innovation or process innovation, but it is also influenced by the technological efficiency of production process. A Stochastic Frontier Production Model (SFM) is a representative method for estimating the technical efficiency of such production. First, as a result of estimating the production function by setting the output variable as total output or value-added, in both cases, the output increased significantly in all manufacturing sectors as inputs of labor, capital, and intermediate increased. On the other hand, R&D investment has a large impact on output in chemical, electronics, and machinery industries. Next, as a result of estimating the technological efficiency through the production function, when the total output is set as the output variable, the overall average of each sector is 0.8 or more, showing mostly high efficiency. However, when value-added was set, Japan had the highest level in most manufacturing sectors, while other countries were lower than the efficiency of the total output. Comparing the three countries of Korea, China and Japan, Japan showed the highest efficiency in most manufacturing sectors, and Korea was about half or one third of Japan and China was lower than Korea. However, in the food and electronics sectors, China is higher than Korea, indicating that China's production efficiency has greatly improved. As such, Korea is not able to narrow its gap with Japan relatively faster than China's rapid growth. Therefore, various policy supports are needed to promote technology development. In addition, in order to improve manufacturing productivity, it is necessary to shift to an economic structure that can raise technological efficiency as well as technology development.

• Journal of Venture Innovation
• /
• v.4 no.2
• /
• pp.29-42
• /
• 2021

The stress on finding a job is also increasing in a situation where the difficulty in finding a job is aggravating due to the COVID-19 pandemic. In this study, the major satisfaction of college students was subdivided into subject satisfaction and relationship satisfaction, and the relationship between these and job-seeking stress was investigated. In addition, We tried to find out whether there is a difference in the influence relationship between these majors according to their current major, that is, whether they majored in a science, engineering major or a social science major. The population for the study was the students currently enrolled in the 4th grade, and the research sample was obtained from students of H and N universities in the metropolitan area. A total of 220 people were analyzed, 110 people from science and engineering and 110 from social sciences. For analysis, SPSS 24.0 and Process Macro 5.0 were used. The empirical analysis results are as follows. First, subject satisfaction had a negative (-) effect on job-seeking stress. Second, relationship satisfaction also had a significant negative (-) effect on job-seeking stress. Third, there was a significant difference between science, engineering students and social science students in the effect of subject satisfaction on job-seking stress. Fourth, in the effect of relationship satisfaction on job-seeking stress, there was also a significant difference between science, engineering students and social science students. Therefore, the higher the satisfaction with the major you are majoring in, the lower the job-seeking stress, and the extent of this decrease is social science students were larger than science, engineering students. It is necessary to be cautious in generalizing the results of this study, which was made in the context of the COVID-19 pandemic. Based on the empirical analysis results, the academic and practical implications of this study are presented.

• International Journal of Quality Innovation
• /
• v.8 no.3
• /
• pp.123-155
• /
• 2007

This research treated the secondary school teachers as the research scale and the research targets included the following: secondary school (employ unit), center of teacher education, and qualified teachers. In terms of these three groups of people, the topics of exploration focused on the perception of suitable teachers, perception of evaluation standard during the process of teacher examination and the relationship among teacher examination systems. The research was managed by two phases: the first phase referred to literature reorganization, expert interview, the qualities and conditions of suitable teachers, important evaluation standard during the process of teacher examination and teacher examination system used; the second phase included questionnaire survey, employ school acquisition, the perception of centers of teacher education and qualified teachers toward the suitability condition and examination evaluation standard in the first phase and teacher examination system used. This research found out that as to the perception of suitable teachers, through data collection, there were six factors reorganized. The levels of their importance were as follows: education devotion, teaching capacity, class management, capacity to guide special students, capacity to communicate with the parents and the will to undertake administrative works. Noticeably, employee unit and centers of teacher education apparently valued class management more, compared with trained teachers; as to evaluation standard of examination, the analytical result found out that the perception of three groups were different in terms of the views toward educational works, written examination data, candidate's age, club experience at school and capacity to use multimedia support teaching materials. This research further proposed six suggestions for centers of teacher education and trained teachers: (l) employ schools considerably valued educational devotion; (2) trained teachers tended to neglect the importance of class management; (3) employee unit considerably cared about the new teachers' competence to use multimedia support teaching materials; however, trained teachers did not have the same view; (4) employee unit considerably cared about new teachers' views toward educational works as well as the candidates' ages; (5) generally speaking, trained teachers neglected the importance of club experience at school; (6) the data revealed that written examination data was not relatively important in terms of teacher examination.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2008.05a
• /
• pp.165-168
• /
• 2008

With the development of the information technology in recent years, the innovation of multimedia information technology also has been accelerated. Many fields of the digital image processing technologies as image data compression, recognition. restoration, etc. are now being studied actively. When transmitting and saving digital images, noise would be made, and we are using the image filters to remove the noise. The Image Filter used Digital Image Process basically has a two-dimensional structure. There are two methods of the filter's creation - One is reiterating one dimension, and the other is using an indivisible two-dimension image filter. The space domain method using mask is the latter one. That is inserting the value-multiplied pixels values faced each other when the two-dimension filter overlapped on input image- to the filter value's center position and the same position in the image. The image filter is being used widely along with one-dimension filter, according each noise. Most people are using various median filters to remove the impulse noise. However, in this paper, I suggested a powerful switching median filter and compared with conventional method for verification.

• Clean Technology
• /
• v.19 no.2
• /
• pp.113-120
• /
• 2013

The treatment of chemically stable perflourocompounds (PFCs) requires a large amount of energy. An energy efficient arc plasma system has been developed to overcome such disadvantage. $CF_4$, $SF_6$ and $NF_3$ were injected into the plasma torch directly, and net plasma power was estimated from the measurement of thermal efficiency of the system. Effects of net plasma power, waste gas flow rate and additive gases on the destruction and removal efficiency (DRE) of PFCs were examined. The calculation of thermodynamic equilibrium composition was also conducted to compare with experimental results. The average thermal efficiency was ranged from 60 to 66% with increasing waste gas flow rate, while DRE of PFCs was decreased with increasing gas flow rate. On the other hand, DRE of each PFCs was increased with the increasing input power. Maximum DREs of $CF_4$, $SF_6$ and $NF_3$ were 4%, 15% and 90%, respectively, without reaction gas at the fixed input power and waste gas flow rate of 3 kW and 70 L/min. A rapid increase of DRE was found using hydrogen or oxygen additional gases. Hydrogen was more effective than oxygen to decompose PFCs and to control by-products. The major by-product in the arc plasma process with hydrogen was hydrofluoric acid that is easy to be removed by a wet scrubber. DREs of $CF_4$, $SF_6$ and $NF_3$ were 25%, 39% and 99%, respectively, using hydrogen additional gas at the waste gas flow rate of 100 L/min and the input power of 3 kW.