• Proceedings of the Safety Management and Science Conference
• /
• 2009.11a
• /
• pp.187-201
• /
• 2009

Semiconductor industry is based on equipment industry and timing industry. In particular, semiconductor process is very complex and as semiconductor-chip width tails and is becoming equipment gradually more as a high technology. Equipment operation is primarily engaged in semiconductor manufacturing (engineers and operator) of being conducted by, equipment errors have also been raised. Equipment operational data related to the error of korea occupational safety and health agency were based on data and production engineers involved in the operator's questionnaire was drawn through the error factor. Equipment operating in the error factor of 9 big item and 36 detail item detailed argument based on the errors down, and 9 big item the equipment during operation of the correlation error factor was conducted. Each of the significance level was correlated with the tabulation and analysis. Using the maximum correlation coefficient, the correlation between the error factors to derive the relationship between factors were analyzed. Facility operating with the analysis of error factors (big and detail item) derive a relationship between the model saw. The end of the operation of the facility in operation on the part of the two factors appeared as prevention. Safety aspects and ergonomics aspects of the approach should be guided to the conclusion.

• Journal of Intelligence and Information Systems
• /
• v.29 no.1
• /
• pp.249-263
• /
• 2023

Collaborative Filtering, a representative recommendation system methodology, consists of two approaches: neighbor methods and latent factor models. Among these, the latent factor model using matrix factorization decomposes the user-item interaction matrix into two lower-dimensional rectangular matrices, predicting the item's rating through the product of these matrices. Due to the factor vectors inferred from rating patterns capturing user and item characteristics, this method is superior in scalability, accuracy, and flexibility compared to neighbor-based methods. However, it has a fundamental drawback: the need to reflect the diversity of preferences of different individuals for items with no ratings. This limitation leads to repetitive and inaccurate recommendations. The Adaptive Deep Latent Factor Model (ADLFM) was developed to address this issue. This model adaptively learns the preferences for each item by using the item description, which provides a detailed summary and explanation of the item. ADLFM takes in item description as input, calculates latent vectors of the user and item, and presents a method that can reflect personal diversity using an attention score. However, due to the requirement of a dataset that includes item descriptions, the domain that can apply ADLFM is limited, resulting in generalization limitations. This study proposes a Generalized Adaptive Deep Latent Factor Recommendation Model, G-ADLFRM, to improve the limitations of ADLFM. Firstly, we use item ID, commonly used in recommendation systems, as input instead of the item description. Additionally, we apply improved deep learning model structures such as Self-Attention, Multi-head Attention, and Multi-Conv1D. We conducted experiments on various datasets with input and model structure changes. The results showed that when only the input was changed, MAE increased slightly compared to ADLFM due to accompanying information loss, resulting in decreased recommendation performance. However, the average learning speed per epoch significantly improved as the amount of information to be processed decreased. When both the input and the model structure were changed, the best-performing Multi-Conv1d structure showed similar performance to ADLFM, sufficiently counteracting the information loss caused by the input change. We conclude that G-ADLFRM is a new, lightweight, and generalizable model that maintains the performance of the existing ADLFM while enabling fast learning and inference.

• Korean Journal of Child Studies
• /
• v.19 no.2
• /
• pp.77-96
• /
• 1998

The purpose of this study was the development of a scale to assess the daily hassles of school age children. The subjects were 398 third and sixth grade children selected from elementary schools in Seoul, Inchon, Taejon, Kwangju and Pusan. Instruments included a questionnaire composed of a 96-item Daily Hassles Coping Scale, demographic questions, a 27-item Children's Depression Inventory, the 42-item Han & Yoo Daily Hassles Scale for Children, and a 30-item Children's Behavior Check List for mothers. Mean, %, ${\chi}^2$, Cramer's V, Cronbach's ${\alpha}$, factor analysis and Pearson's r were used for data analysis. 82 items from the 96-item Daily Hassles Coping Scale were selected through the item discriminant method. (Cramer's V discriminant coefficients ranged from .36 to .79). Five factors were extracted from the lst and 2nd factor analyses. Subscales were "parent-related," "friend-related I," "study-related," "friend-related II," and "teacher-related" stress. Criterion validity of the 82-item scale was confirmed through significant correlations with the Children's Depression Inventory (r=.55) and the Han & Yoo Daily Hassles Scale for Children (r=.22). Internal consistency of this scale was high, including internal reliability of subscales. These results confirm this scale as a valid and reliable measure of daily hassles of school age children.

• Korean Journal of Adult Nursing
• /
• v.12 no.4
• /
• pp.629-640
• /
• 2000

The purpose of this study was to develop and validate the scale to measure dementia patient's caregiver burden of Korea. In the first phase of the study, 15 caregivers of dementia patients were interviewed to provide narrative data from which items were developed. Initially 65 items were generated from the interview data of 15 caregivers. Content validity was judged by two separate panels of experts with 27 professionals and 30 family caregivers. These items were analyzed through the Index of Content Validity and 33 items were selected which met .80 or more of the CVI. This preliminary FCBSD-K was tested with 207 adult caregivers for reliability and construct validity including item analysis and orthogonal(Varimax) factor analysis. Eight items were deleted because of high or low item-item correlation. The result of the second factor analysis produced six factors that coincided with the conceptual framework posed for the scale developed. The six factors were labeled as 'physio social factor' 'emotional factor' 'family cultural factor' 'role obligation' 'guilt feeling' and 'financial & supportive system factor'. The alpha coefficient relating to internal consistency was .9264 for reliability. In conclusion, cultural factor is related to dementia patient's caregiver burden and FCBSD-K was useful in assessing the dementia patient's caregiver burden in Korea.

• Journal of Korean Elementary Science Education
• /
• v.35 no.2
• /
• pp.137-149
• /
• 2016

The purpose of this study is to develop a descriptive paper test item which can evaluate elementary school students' HGA (scientific Hypothesis Generating Ability) and to propose a counting formula that can easily assess student's HGA objectively and quantitatively. To make the test item can possibly evaluate all the students from 6th graders to 3rd graders, the 'rabbit's ear' item is developed. Developed test item was distributed to four different elementary schools in Seoul. Total 280 students who were in the 6th grade solved the item. All the students' reponses to the item were analyzed. Based on the analyzed data evaluation factors and evaluation criteria are extracted to design a Hypothesis Generating ability Quotient (HGQ). As the result 'Explican's Degree of Likeness' and 'Hypothesis' Degree of Explanation' are chosen as evaluation factors. Also precedent evaluation criteria were renewed. At first, Explican's Degree of Likeness evaluation criterion was turned four levels into three levels and each content of evaluation criterion is also modified. Secondly, new evaluation factor 'Hypothesis' Degree of Explanation' was developed as combined three different evaluation criteria, 'level of explican', 'number of explican' and 'structure of explican'. This evaluation factor was designed to assess how the suggested hypothesis can elaborately explain the cause of one phenomenon. Newly designed evaluation factors and evaluation criteria can assess HGA more in detail and reduce the scoring discordant through the markers. Lastly, Developed counting formula is much more simple than precedent Kwon's equation for evaluating the Hypothesis Explanation Quotient. So it could help easily distinguish one student's scientific hypothesis generating ability.

• Journal of Technologic Dentistry
• /
• v.10 no.1
• /
• pp.159-169
• /
• 1988

Fixed prosthetic appliance is the restorations of damaged teeth with cast metal or porcelain, and of replacing missing teeth with fixed, or cemented prosthesis. Successfully treating a patient by means of fixed prosthodontics requires the thoughtful combination of dentist and dental laboratory technician. In order to know the satisfaction of dentisits about fixed prosthetic appliances-porcelain, precious and non precious metal-, this study was conducted for 69 practicing dentisis in Pusan during Fed. to Mar. 1988 using a questionnaire method and was analyzed by the use of percentage, $X^2$-test The results are obtained as follow: 1. General characteristics of the respondents, The rate of age distribution of practcing dentists responded was as follow : 31$\sim$35 age group is 29.0%, 51$\sim$55age group 18.8%, 41$\sim$45 age group and 46$\sim$50 group each 11.6%, 36$\sim$40 age group and 56$\sim$60 age group 8.7%, over 60 age group 7.3%, below 30 age group 4.3% respectively. The tate of term of practice experience distribution of dentists responded was as follow : 6$\sim$10 year group 26.1%, 25$\sim$25year group 18.8%, 11$\sim$15 years group and 16~20 years group each 13.0%, below 5 years group and over 30 years group each 10.2%, 26$\sim$30 years group 8.7% in order. The rate of respondents’ degree was follow : bachelor group 81.2%, doctor group 13.0%, master group 5.8% in order. The rate of respondents’completed medical specialist course was as follow : non-complete group 75.4%, complete group 24.6%. The rate of having own laboratory in respondents’clinic was as follow : no having own laboratory group 58.0%, having own labiratory group 42.0% The rate of being lab. technician in respondent;s clinic was as follow: non-being lab. technician group 91.3%, being lab. technician group 8.7%. The rate of transactional commercial dental lab. numbers was as follow : I dental lab. group 40.6%, 2 dental lab. group 30.4%, 3 dental lab. group 20.3%, production at own lab. group 8.7% 2. Cognitions about the respondent's job, About cognition of patient numvers, there were neither many nor few group 62.3%, few group 17.4%, many group 15.9%, too many group 4.4% in order.(P<.01) As compared with the ratio of dentists to patient prothetic treatment, there were moderate level group 46.4%, high level group 33.6%, low level group 18.8% in order.(P<.01) By job satisfaction of respondents, the highest percentage group was satisfied group 44.9%, neither astisfied nor disstified group 33.3%, dissatisfied group 14.5%, much satisfied group 7.3%(P<.01). As compared with the level of prosthetics price, the were moderate level group 71.0%, low level group 23.2%, high level group 5.8%(P<.01) 3. Satisfaction and assessment of dissatisfied factor of the fixed prosthetic appliance, As compared with the satisfaction of al prosthetic appliance, there were neither satisfied nor dissatisfied group 59.4%, satisfied group 20.3% dissatisfied group 18.8%, much satisfied group 1.5% in order.(P<.01). About precious and non precious metal crown and bridge, there were neither satisfied nor dissatisfied group 60.9%, satisfied group 24.6%, dissatisfied group 11.6%, much satisfied group 2.9% in order.(P<.01) The reat of respondent;s satisfaction about porcelain crown and bridge was as follow : neither satisfied nor dissatisfied group 43.5%, dissatisfied group 33.3%, satisfied group 21.7%, much satisfied group 1.5%(P<.01). The rate of respondent's dissatisfied factor item of precious and non precious metal crown and bridge was as follow : inadequate margin fit item 39.1% respondent, inadequate occlusion item 24.6%, inadequate adaptation item 23.2%, poor anat omic form item 15.9%, inadequate contact item 31.9%, improper polishing item 4.4%, the rest item 11.6% respectively. The rate of respondent's dissatisfied factor item of porcelain crown and bridge was as follow : inadequate margin fit item 52.2%, inadequate occlusion item 2.9%, inadequate adaptation item 10.1%, poor anatomic form item 11.6%, improper polishing item 1.5%, shade mismatching item 71%, the rest item 7.2% respectively. The rate of cognition of respondents’ about cause of dissatisfied factor on fixed prosthetic appliance by item was as follow : deficiency of sincerity item 55.1%, deficiency of knowledge about work item 47.8%, deficiency of experience item 15.9%, poor quality of using material item 5.8%, the rest item 2.9% respectively.

• Journal of the Korean Home Economics Association
• /
• v.39 no.1
• /
• pp.221-232
• /
• 2001

The purpose of this study was to develop a scale for evaluation of preschool children's peer competence and to examine its validity and reliability. The 15 items of the preliminary scale were completed by teachers, and peer rating scale for assessment of peer acceptance was administered to 365 children 5 to 6 years of age. The methods for data analysis included item analysis, factor analysis for construct validity, Pearson correlations between teacher's ratings of peer competence and children's ratings of peer acceptance for concurrent validity, and Cronbach's u for reliability. The 15 items of the scathe were found be satisfactory in terms of item response distribution and item discrimination. Factor analysis revealed that a 3-factor solution was the best fit: sociability, prosocial behavior, leadership in peer relations. Correlations between teacher's ratings and children's ratings were as significant. Cronbach'so ranged from .88 to .93 for three subscales and .93 for the total scale. It was concluded that the psychometric properties of the peer competence scale in general were acceptable for use.

• Journal of the Korea Society of Computer and Information
• /
• v.25 no.6
• /
• pp.199-206
• /
• 2020

This study was conducted for the purpose of developing a measurement tool for success and failure factors of education and training of Korean bodyguards. conducted a meeting from the fully open questionnaire at first, and then formed the semi-structured questionnaire, finally carried out the survey from the closed questionnaire and analyzed data from SPSS 21.0, AMOS 21.0 and developed the measurements. It was conducted from May, 2019 to December, 2019. This survey was conducted of 150 security guards after the verification of the content validity though the pilot survey and presented the success attribution factors and standards on the basis of the result form this survey. As a result, the success factors of the training of the bodyguards were accidental education (5 item), vocational mental education (2 item), vocational mental education (2 item), work ability enhancement education (2 item), realistic practical education (2 item) ), Including 4 items, 11 items, The failure factors consisted of 12 item of three factors: formal education and training (5 item), lack of leadership qualities (4 item), and lack of education (3 item).

• Child Health Nursing Research
• /
• v.8 no.3
• /
• pp.291-301
• /
• 2002

This study was performed to develop a scale of maternal conflict based on already established scales and the literature review. The major purpose of this study was to develop a scale of which contains difficulties which mother can feel to rearing her child and can be applied to mothers at home or employed mothers. 68 items were developed. For content validity test, the items were examined by 7 experts panel with Likert 4 scale. The 39 items were selected as high validity from 86% of the experts. The item number of each area and the misunderstanding words were corrected by advices of experts. 10 mothers again examined 39 items in order to correct the misunderstanding words and the long sentences. For reliability and factor analysis of 39 items, 496 mothers participated. 34 items were chosen because 5 items that corrected item-total correlation coefficient were lower than 0.2 were excluded by the result of reliability analysis. Factor analysis was done with 34items. 6 factors were classified higher than eigen value 1.0, 28 items were chosen because 6 items that factor loadings were lower than 0.4 were excluded. These factors were named as employment and caring for child(6 items), child education(5 items), support system for caring for child(4 items), interaction with mother and child(4 items), view of maternal role(5 items) and me as human being(4 items). Reliability analysis was done with 28 items. 1 item(36 number of factor 1) was excluded because corrected item-total correlation coefficient were lower than 0.2. Therefore, 27 items were chosen. Maternal Conflict Scale was contained 27 items. The corrected item-total correlation coefficient of 27 items were ranged from .2345 to .5240 and Cronbach alpha coefficient of those was .8207 by the results of reliability analysis. It can be evaluated that this scale had a high validity and reliability because of passing through an objective process of test.

• Korean Journal of Social Welfare
• /
• v.59 no.4
• /
• pp.273-296
• /
• 2007

In this Study, to reconstruct the disability identity scale(Lee and Shin, 2006)), Rasch rating scale model was applied to the four sub-dimensions of the Disability Identity Scale in a sample of spinal cord injuries(N=397). The Disability Identity Scale was verified by explorative factor analysis and confirmatory factor analysis. However, factor analytic procedures can't evaluate item-fit indices, item difficulty, and appropriate scale category. A number of limitations posed by confirmatory factor analytic procedures can be averted with the use of Rasch rating scale model which is in the item response theory(IRT). So in this study, Rasch model was applied to the Disability Identity Scale. Results revealed that (A) 20 items were selected from Rasch model, (B) the difficulty level of the Disability Identity Scale was the average level, (C) 4-point rating scale was appropriate for the Disability Identity Scale. Finally, we could suggest that the sub-dimensions concepts of the disability identity became clearer and items were to the good fitting.