• Journal of Family Resource Management and Policy Review
• /
• v.25 no.1
• /
• pp.91-112
• /
• 2021

This study aimed to explore the influence of family characteristics on the work-life balance of early adolescents. A series of data analyses was conducted on adolescents' use of time in daily life on the basis of 2018 Korean Children and Youth Panel Survey (KCYPS 2018). We found that the work-life balance of youth is related to their parents' health status, presence of older siblings, household income, parenting attitude, parent-child communication time, and mother's occupation. The work-life balance of the fifth graders is more likely to be influenced by family characteristics compared with that of the eighth graders. In particular, the fifth graders' sleep deprivation is affected by the mother's occupation, but there is no significant effect on the eighth graders' sleep deprivation. An important factor in skipping breakfast is household income, with adolescents from low-income families tending to skip breakfast more than five days a week. In addition, parents' health status and parenting attitude are significantly related to skipping of breakfast in early adolescents. Household income is related to the after-school private tutoring hours of both the fifth and eighth graders; however, parenting attitude and mother's occupation are also significant influencing factors of the fifth graders' after-school study. Mother's occupation is related to excessive cell phone use; specifically, the fifth graders whose mothers work white-collar jobs, sales and services or manufacturing are more likely to play with cell phones more than three hours a day than those whose mothers are full-time housewives. These results suggest that the work-life balance policies targeted at the family characteristics of adolescents can improve family environments in a manner that enhances adolescents' life balance, thus supporting the well-being of early adolescents and their families.

• Journal of Korean Home Economics Education Association
• /
• v.33 no.3
• /
• pp.41-63
• /
• 2021

The purpose of this study was to investigate the nutritional knowledge and eating behavior of middle school students in Gwangju area in order to provide basic data for the development of home economics curriculum that can help the students form healthy eating habits. For this purpose, a self-administered questionnaire was distributed to middle school students in Gwangju area. A total of 330 questionnaires were collected by convenience sampling and analyzed using SPSS(Statistics Package for the Social Science) Version 20.0 for Windows. Results of this study are as follows. First, school curriculum session ranked top(31.82%) on the list of sources for middle school students to acquire nutrition knowledge. Second, the mean score of nutrition knowledge of all respondents was moderately high(14.33 points out of maximum 20 points). In terms of nutrition knowledge by gender, female students had a higher level of nutrition knowledge in the 'Food' domain than their male counterparts(p<0.05). With regard to nutrition knowledge by grade, Grade 9 showed a higher score than Grade 7 and 8(p<0.001). Third, the mean score of eating behavior of all respondents was at an average level(69.75 out of 100 points). In terms of eating behavior by gender, male students showed a higher score than their female counterparts(p<0.05). In particular, male students had higher scores than female students for the following items: "I exercise regularly after school"(p<0.001); "I regularly eat meal three times per day"(p<0.01); "I don't skip breakfast"(p<0.01); and "I don't eat sweet food often"(p<0.01). In terms of eating behavior by grade, Grade 9 showed higher scores than Grades 7 and 8 for the following items: "I eat meal out of thankfulness for those who have prepared food"(p<0.01) and "I eat grains for every meal"(p<0.01). Finally, with regard to eating behavior depending on the level of nutrition knowledge, the 'Upper' and 'Middle' groups had higher scores for eating behavior than the 'Lower' group, indicating that a lower level of nutrition knowledge resulted in a lower score in eating behavior. Based on the above results, home economics teachers responsible for dietary education should have a greater sense of mission and pride and make more efforts to improve nutrition knowledge and eating behavior of middle school students.

• Journal of Intelligence and Information Systems
• /
• v.28 no.2
• /
• pp.127-146
• /
• 2022

Recently, as word embedding has shown excellent performance in various tasks of deep learning-based natural language processing, researches on the advancement and application of word, sentence, and document embedding are being actively conducted. Among them, cross-language transfer, which enables semantic exchange between different languages, is growing simultaneously with the development of embedding models. Academia's interests in vector alignment are growing with the expectation that it can be applied to various embedding-based analysis. In particular, vector alignment is expected to be applied to mapping between specialized domains and generalized domains. In other words, it is expected that it will be possible to map the vocabulary of specialized fields such as R&D, medicine, and law into the space of the pre-trained language model learned with huge volume of general-purpose documents, or provide a clue for mapping vocabulary between mutually different specialized fields. However, since linear-based vector alignment which has been mainly studied in academia basically assumes statistical linearity, it tends to simplify the vector space. This essentially assumes that different types of vector spaces are geometrically similar, which yields a limitation that it causes inevitable distortion in the alignment process. To overcome this limitation, we propose a deep learning-based vector alignment methodology that effectively learns the nonlinearity of data. The proposed methodology consists of sequential learning of a skip-connected autoencoder and a regression model to align the specialized word embedding expressed in each space to the general embedding space. Finally, through the inference of the two trained models, the specialized vocabulary can be aligned in the general space. To verify the performance of the proposed methodology, an experiment was performed on a total of 77,578 documents in the field of 'health care' among national R&D tasks performed from 2011 to 2020. As a result, it was confirmed that the proposed methodology showed superior performance in terms of cosine similarity compared to the existing linear vector alignment.