A Cost-Effective CNN-LSTM-Based Solution for Predicting Faulty Remote Water Meter Reading Devices in AMI Systems.

Automatic meter infrastructure (AMI) systems using remote metering are being widely used to utilize water resources efficiently and minimize non-revenue water. We propose a convolutional neural network-long short-term memory network (CNN-LSTM)-based solution that can predict faulty remote water meter reading (RWMR) devices by analyzing approximately 2,850,000 AMI data collected from 2762 customers over 360 days in a small-sized city in South Korea. The AMI data used in this study is a challenging, highly unbalanced real-world dataset with limited features. First, we perform extensive preprocessing steps and extract meaningful features for handling this challenging dataset with limited features. Next, we select important features that have a higher influence on the classifier using a recursive feature elimination method. Finally, we apply the CNN-LSTM model for predicting faulty RWMR devices. We also propose an efficient training method for ML models to learn the unbalanced real-world AMI dataset. A cost-effective threshold for evaluating the performance of ML models is proposed by considering the mispredictions of ML models as well as the cost. Our experimental results show that an F-measure of 0.82 and MCC of 0.83 are obtained when the CNN-LSTM model is used for prediction.

Contextualizing geneticization and medical pluralism: How variable institutionalization of traditional, complementary, and alternative medicine (TCAM) conditions effects of genetic beliefs on utilization.

As genetic models are increasingly incorporated in medicine, health service users seem to accept these models to varying degrees. To appreciate these differences, this paper examines how health service users' genetic beliefs of health are associated with their use of traditional, complementary, and alternative medicine (TCAM) via responses from 31 countries in the 2011 ISSP survey. It finds an interesting contrast between East Asian countries and other countries in the world. The negative association between genetic beliefs and TCAM use is strong in the latter, whereas it is weak in the former. More intriguingly, the analysis demonstrates significant cross-national differences within East Asian countries. Chinese and Koreans reveal a negative relationship between genetic beliefs and TCAM use, while Japanese show a positive relationship. The paper provides an explanation to these cross-national differences by drawing on comparative studies of medical systems. When TCAM is institutionalized as valid and distinct medical resources, and when TCAM is practically available to everyday use, health service users who subscribe to genetic beliefs are likely to use TCAM as well. These findings contribute to revealing the pragmatic nature of health-seeking action and the institutional context in which geneticization and medical pluralism are conditioned to form.

RN+: A Novel Biclustering Algorithm for Analysis of Gene Expression Data Using Protein-Protein Interaction Network.

Biclustering is a process of finding groups of genes that behave similarly under a subset of conditions. In this article, we propose an efficient biclustering algorithm, namely RN+, to identify biologically meaningful biclusters in gene expression data. The RN+ algorithm finds biologically meaningful biclusters through a novel gene filtering using protein-protein interaction network, gene searching, gene grouping, and queuing process. It also efficiently removes duplicate biclusters. We tested the proposed RN+ on five real microarray datasets, and compared its performance with seven competitive biclustering algorithms. The experimental results show that RN+ efficiently finds functionally enriched and biologically meaningful biclusters for large gene expression datasets, and outperforms the other tested biclustering algorithms on real datasets.

Cross-national differences in the holistic use of traditional East Asian medicine in East Asia.

Complementary and alternative medicine (CAM) has been one of the popular strategies for health promotion. Traditional East Asian medicine (TEAM) is one of the most popular CAM practices in the world and there are suggestions that its holistic utilization is important for users to gain its effects for health promotion. In this context, this study investigates the extent to which TEAM users in East Asian countries utilize various modalities of TEAM holistically. It provides a model that explains cross-national differences in the extent of the holistic use of TEAM between China, Japan, Korea and Taiwan. Using the 2010 East Asian Social Survey, regression models specify the relationship between the holistic use of TEAM and the geographical location (country). The presence of TEAM doctors who hold the comprehensive and exclusive practice rights over TEAM is found to be conducive to the holistic utilization of various TEAM modalities. Thus, Taiwanese and Koreans use TEAM more holistically than Chinese and even more so than Japanese. The result suggests that the manner in which TEAM is institutionalized affect the extent to which TEAM users utilize various TEAM modalities together and potentially the health promotion effects of TEAM.

Marital status and mortality: Does family structure in childhood matter?

It is well known that marital status is significantly associated with mortality risk. Little is known, however, regarding whether and how the effects of marital status are moderated by one's own family structure in childhood. The purposes of this study are to examine whether marital status (i.e., family structure in adulthood) and living with both biological parents in childhood (i.e., family structure in childhood) are associated with mortality risk, and whether and how the effects of marital status vary depending on family structure in childhood and gender. We analyze the risk of death in five waves of the General Social Survey (GSS) from 1994 through 2002 after linking the GSS data to death certificate data from the National Death Index through 2008. The findings indicate that being widowed increases the risk of mortality, while living with both parents in childhood lowers it. Interestingly, analysis of the interaction between marital status and family structure in childhood reveals that the disadvantage of widowhood in terms of mortality is significantly stronger for those who lived with both parents in childhood than for those who did not. Subsample analysis by gender shows that the moderating effect of living with both parents is largely equal across men and women, though statistically more robust for men. These findings suggest that living with both parents during childhood may increase vulnerability to marital disruptions due to unwanted life events such as spousal loss. Childhood advantages, ironically, may form more stressful contexts of spousal loss by lowering one's adaptability or immunity to adulthood hardships, especially when the hardships in adulthood are characteristically opposite from the childhood advantages.

Living arrangements and suicidal ideation among the Korean older adults.

OBJECTIVES: This study examines how living arrangements are associated with suicidal ideation for older adults in South Korea, which has the highest suicide rate among OECD countries, and a particularly high suicide rate for older persons. METHODS: Analyzing a sample of 5795 women and 3758 men aged 65 and older from a nationwide representative cross-sectional data-set, we examined how many older adults think about suicide over a one-year period, why they think about suicide, and whether living arrangements are associated with suicidal ideation. RESULTS: About 1 out of 12 respondents in our sample reported suicidal ideation. While women and men did not differ in the prevalence of suicidal ideation, women attributed their suicidal feelings to health problems, while men attributed theirs to economic difficulties. Logistic regression results indicated that living arrangements are associated with suicidal ideation for men but not women. Older men living with a spouse were less likely to have suicidal ideation than older men with other living arrangements (i.e., living alone, living with children without spouse, living with spouse, and others). CONCLUSIONS: Our results highlight the importance of living arrangements to older men's suicidal ideation. We discuss gender differences in the implications of living arrangements to suicidal ideation within the context of Confucian culture.

Religious Affiliation, Religious Service Attendance, and Mortality.

Very few studies have examined the effects of both religious affiliation and religiosity on mortality at the same time, and studies employing multiple dimensions of religiosity other than religious attendance are rare. Using the newly created General Social Survey-National Death Index data, our report contributes to the religion and mortality literature by examining religious affiliation and religiosity at the same time. Compared to Mainline Protestants, Catholics, Jews, and other religious groups have lower risk of death, but Black Protestants, Evangelical Protestants, and even those with no religious affiliation are not different from Mainline Protestants. While our study is consistent with previous findings that religious attendance leads to a reduction in mortality, we did not find other religious measures, such as strength of religious affiliation, frequency of praying, belief in an afterlife, and belief in God to be associated with mortality. We also find interaction effects between religious affiliation and attendance. The lowest mortality of Jews and other religious groups is more apparent for those with lower religious attendance. Thus, our result may emphasize the need for other research to focus on the effects of religious group and religious attendance on mortality at the same time.

The general social survey-national death index: an innovative new dataset for the social sciences.

BACKGROUND: Social epidemiology seeks in part to understand how social factors--ideas, beliefs, attitudes, actions, and social connections--influence health. However, national health datasets have not kept up with the evolving needs of this cutting-edge area in public health. Sociological datasets that do contain such information, in turn, provide limited health information. FINDINGS: Our team has prospectively linked three decades of General Social Survey data to mortality information through 2008 via the National Death Index. In this paper, we describe the sample, the core elements of the dataset, and analytical considerations. CONCLUSIONS: The General Social Survey-National Death Index (GSS-NDI), to be released publicly in October 2011, will help shape the future of social epidemiology and other frontier areas of public health research.

Volunteering among older people in Korea.

OBJECTIVE: Faced with aging societies, there is an immense need to better understand the nature of volunteering outside advanced Western industrial countries. As a case of a rapidly aging society, we identify robust factors associated with elderly volunteering in Korea in terms of a resource framework. METHODS: Data were derived from the Social Statistics Survey conducted by the Korea National Statistical Office in 1999 (N = 7,135) and 2003 (N = 8,371). We first determined overall and age-related volunteer rates for Korea compared to the United States. Using logistic regression, we then examined the effects of human, cultural, and social capital variables on volunteering. RESULT: Approximately 6% of Koreans aged 65 years and older participate in volunteer programs. All human capital variables are positively related with volunteering. For cultural capital, those who identify their religion as Buddhism or Catholicism are more likely to volunteer than those who have no religion. But surprisingly, Protestantism does not consistently promote volunteering across both years. For social capital, older adults who live alone or with a spouse are more likely to volunteer than those living with both a spouse and children. DISCUSSION: In contrast to human capital, cultural and social capital on elderly volunteering appears to be contoured by social contexts.