• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.11a
• /
• pp.1230-1233
• /
• 2023

최근, 65세 이상의 1인 가구가 급증함에 따라 노인을 대상으로 한 다양한 연구 및 서비스가 활발히 이루어지고 있다. 이에 본 논문에서는 시계열 센서 데이터를 이용하여 CNN-LSTM 기반의 노인 활동 인식 시스템을 제안한다. 수집된 데이터는 3축 가속도 센서가 내장된 2개의 디바이스를 등과 허벅지에 부착하였다. 수집 주기는 50hz로 진행되었으며, 각 행동은 2초를 기준으로 산정하였다. 학습데이터의 입력값으로 사용하기 위해, 슬라이딩 윈도우를 50%로 적용하여 시퀀스를 구성하였다. 모델은 특징을 반영하기 위한 CNN(Convolutional Neural Networks)과 시계열적 특성을 반영하기 위한 LSTM(Long-Short Term Memory)을 하이브리드한 1차원 형태의 CNN-LSTM 모델을 사용한다. 행동은 4가지로 분류하였으며, 97%의 정확도를 나타내고 있다.

• 한국사회복지학회:학술대회논문집
• /
• 2007.04a
• /
• pp.377-381
• /
• 2007

• Communications of the Korean Institute of Information Scientists and Engineers
• /
• v.32 no.11
• /
• pp.16-20
• /
• 2014

• The Journal of the Korea Contents Association
• /
• v.14 no.11
• /
• pp.328-336
• /
• 2014

This work has performed to find what activities of daily living are required for the intensive interests when inpatient elderly more than 3 months has been supported and convalescent care, where the inpatient elderly were judged by the inpatient assessment report in the time of December, 2013. According to the estimation with logistic function of the relationship between the state of decision making recognition technology and the Activities of Daily Living(ADL), the intensive cares for the elderly are required in the parameters of 'Having meal' and 'transferring sitting' when they are severed and convalescently cared as the degree of functional independence for ADL are severly proceeded. In addition, the senescence and disease the activities except 'Having meal' and 'transferring sitting' seem to be influenced by the decline of body function more than the state of decision making recognition technology for daily living.

• Korean Journal of Social Welfare
• /
• v.59 no.3
• /
• pp.175-199
• /
• 2007

The study analyzes the characteristics, work stress, satisfaction and attitudes toward dementia of social care work force for dependent elders in Korea. Data were from 502 staffs including social workers, nurses, and direct care workers from 45 diverse type of nursing facilities in Busan and Daegu area. The results of the analysis indicate that they are low paid and overloaded in general. The respondents espoused highly hopeful and person-centered attitudes toward dementia and the elderly, while they showed low level of satisfaction related to work. The stress levels were higher in sub scales concerning care tasks and physical environments in work places. In addition, the results of multiple regression denote that those have higher level of satisfaction who are nurses, have no intension to quit, have more experiences of work education, and working in facilities with more frail elders. Stresses were closely related to higher level of education and the intension to quit. Person centered attitude was more often reported by those who have more elderly clients to take care of and are working in facilities with more elders who are demented and over 80. In particular, the association was consistent between higher level of job satisfaction and the person centered attitude. Several practical suggestions linked to the analysis were made including improving the welfare for staffs working in nursing facilities and providing continuous professional training and education for them particularly on person-centered care. In addition, it was emphasized to raise the morale of social care work force considering the rapidly increasing need of long term care and the important influence that care work force has on older persons' quality of life from now on.

• Journal of Convergence for Information Technology
• /
• v.11 no.9
• /
• pp.43-49
• /
• 2021

This study was conducted to investigate the correlation between diabetes empowerment, health perception, and diabetes self-care behaviors with diabetes for the elderly in the community, and to identify the factors affecting diabetes self-care behavior in the elderly with diabetes. A survey was conducted on 80 diabetic elderly people over the age of 65 who were registered at a local public health center. The mean age was 71.15 years, and 41 males and 39 females were included. Diabetes self-care behavior showed a significant positive correlation with economic status (r=.245, p=.029) and diabetes empowerment (r=.406, p<.001), but health perception (r=.127, p=.263) did not show a significant correlation. As a result of this study, diabetes empowerment and economic status of the diabetic elderly were found to be significant influencing factors on diabetes self-care behavior in the diabetic elderly, and the explanatory power of the model was 19.6% (F=10.623, p<.001). Therefore, if a community program is developed to improve the diabetes empowerment of the elderly with diabetes in the community and economic support policies are also provided at the level of public health, it will be possible to improve the self-care behavior of the elderly with diabetes in the community.

• The Journal of the Korea Contents Association
• /
• v.21 no.4
• /
• pp.576-590
• /
• 2021

As a 'Senior Citizen' by applying an active theory based on a role model, elderly volunteers may seek to impart health to the local communities through volunteer activities. In addition, the strategies for promoting the elderly participation volunteering actions may apply with the lifelong education/learning model for citizenship development, which intends to expand the elder volunteer activity participants and the areas of volunteering actions. In this study, we attempted to find out how volunteering fieldworkers experience and perceive volunteer-promoting strategies for senior volunteers drawn from previous literatures and studies on promoting elderly volunteer actions. For the purpose of the study, we conducted in-depth interviews with 12 fieldworkers who worked senior-welfare centers and volunteer centers in Seoul and Kyeong-gi area. We used the directed-content analysis, which categorized 12 themes/categories commonly mentioned from previous literatures and studies. Within these categories, the analysis was made to derive issues and improve promoting strategies specified from the interviewees. The outcomes of the study included the insights regarding what promoting-strategies might be necessary to enrich existing senior volunteer actions. The study highlighted not only reviews volunteer fieldworkers' current experiences with senior volunteers but also barrier and practical suggestions for the continuous advances of senior volunteer programs and strategies.

• Journal of Internet of Things and Convergence
• /
• v.8 no.4
• /
• pp.25-40
• /
• 2022

This study aimed to understand the increasing number of elder abuses in South Korea, where entry into the super-aged society is imminent, by implementing text mining analysis. Korean Academic journals were obtained from 2004, the establishment year of the senior care agency, to 2021. We performed natural language processing of the titles, keywords, and abstracts and divided them into three segments of periods to identify latent meanings in the data. The results illustrated that the first section included 81 papers, the second 64, and the third 104 respectively, averaging 13.8 annually, which increased its numbers from 2014 until the decrease below the annual average in 2020. Word frequency demonstrated that the common keywords of the entire segments were 'elder abuse,' 'elders,' 'influences,' 'factors,' 'recognition,' 'family,' 'society,' 'prevention plans,' 'experiences,' 'abused elders,' 'abuse prevention,' 'depression,' etc., in consecutive order. TF-IDF indicated that 'influences,' 'recognition,' 'society,' 'prevention plans,' 'abuse prevention,' 'experiences,' 'depression,' etc., were the common keywords of all divisions. Network text analysis displayed that the commonly represented keywords were 'elder abuse,' 'elders,' 'influences,' 'factors,' 'characteristics,' 'recognition,' 'family,' 'prevention plans,' 'society,' 'abuse prevention,' and 'experiences' in the entire sections. concor analysis presented that the first segment consisted of 5 groups, the second 7, and the third 6. We suggest future directions for elder abuse research based on the results.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.17 no.10
• /
• pp.509-519
• /
• 2016

This study intends to examine how the subjective health perception of drinkers influences the depression trajectory change by using longitudinal data for elderly drinkers. This study has set up a balanced panel by combining the results from "Aging study panel research" conducted in the year 2006 (1st), 2008 (2nd), 2010 (3rd), 2012 (4th). A total of 543 elderly drinkers, who have responded to each research have been selected as targets for the final analysis. In the analyses, descriptive statistics and Latent Growth Modeling were conducted to examine the causal relationship between the observed and latent variables. The results are as follows: First, it was found that there was a negative correlation-with statistical significance-between the initial value of subjective health perception and the initial status of depression symptom. Second, there was a positive correlation between the initial value of subjective health perception and the changes of depression symptom. This indicates that elderly drinkers with high health perception usually have high degree of depression change; however, this does not mean it was statistically significant. Third, there was a statistically significant correlation between subjective health perception change and depression symptom change. It was found that the depression change level would be low if the subjective health perception change level was high. Thus, we can assume that the depression symptoms of elderly drinker c would slow down if the subjective health perception level increases.

• The Korean Journal of Applied Statistics
• /
• v.36 no.1
• /
• pp.33-48
• /
• 2023

In this paper we consider automatic speech recognition (ASR) for Korean speech data in which elderly persons randomly speak a sequence of words such as animals and vegetables for one minute. Most of the speakers are over 60 years old and some of them are dementia patients. The goal is to compare deep-learning based ASR models for such data and to find models with good performance. ASR is a technology that can recognize spoken words and convert them into written text by computers. Recently, many deep-learning models with good performance have been developed for ASR. Training data for such models are mostly composed of the form of sentences. Furthermore, the speakers in the data should be able to pronounce accurately in most cases. However, in our data, most of the speakers are over the age of 60 and often have incorrect pronunciation. Also, it is Korean speech data in which speakers randomly say series of words, not sentences, for one minute. Therefore, pre-trained models based on typical training data may not be suitable for our data, and hence we train deep-learning based ASR models from scratch using our data. We also apply some data augmentation methods due to small data size.