The application of an ensemble of boosted elman networks to time series prediction : a benchmark study

In this paper, the application of multiple Elman neural networks to time series data regression problems is studied. An ensemble of Elman networks is formed by boosting to enhance the performance of the individual networks. A modified version of the AdaBoost algorithm is employed to integrate the predictions from multiple networks. Two benchmark time series data sets, i.e., the Sunspot and Box-Jenkins gas furnace problems, are used to assess the effectiveness of the proposed system. The simulation results reveal that an ensemble of boosted Elman networks can achieve a higher degree of generalization as well as performance than that of the individual networks. The results are compared with those from other learning systems, and implications of the performance are discussed.

Is a perceived supportive physical environment important for self-reported leisure time physical activity among socioeconomically disadvantaged women with poor psychosocial characteristics? An observational study

Background Over the past decade, studies and public health interventions that target the physical environment as an avenue for promoting physical activity have increased in number. While it appears that a supportive physical environment has a role to play in promoting physical activity, social-ecological models emphasise the importance of considering other multiple levels of influence on behaviour, including individual (e.g. self-efficacy, intentions, enjoyment) and social (e.g. social support, access to childcare) factors (psychosocial factors). However, not everyone has these physical activity-promoting psychosocial characteristics; it remains unclear what contribution the environment makes to physical activity among these groups. This study aimed to examine the association between the perceived physical environment and self-reported leisure-time physical activity (LTPA) among women living in socioeconomically disadvantaged areas demonstrating different psychosocial characteristics.

Methods In 2007–8, 3765 women (18–45 years) randomly selected from low socioeconomic areas in Victoria, Australia, self-reported LTPA, and individual, social and physical environmental factors hypothesised within a social-ecological framework to influence LTPA. Psychosocial and environment scores were created. Associations between environment scores and categories of LTPA (overall and stratified by thirds of perceived environment scores) were examined using generalised ordered logistic regression.

Results Women with medium and high perceived environment scores had 20-38% and 44-70% greater odds respectively of achieving higher levels of LTPA than women with low environment scores. When stratified by thirds of psychosocial factor scores, these associations were largely attenuated and mostly became non-significant. However, women with the lowest psychosocial scores but medium or high environment scores had 76% and 58% higher odds respectively of achieving ≥120 minutes/week (vs. <120 minutes/week) LTPA.

Conclusions Acknowledging the cross-sectional study design, the findings suggest that a physical environment perceived to be supportive of physical activity might help women with less favourable psychosocial characteristics achieve moderate amounts of LTPA (i.e. ≥120 minutes/week). This study provides further support for research and public health interventions to target perceptions of the physical environment as a key component of strategies to promote physical activity.

Patron offending and intoxication in night time entertainment districts (POINTED) : a study protocol

Risky alcohol consumption is the subject of considerable community concern in Australia and internationally, particularly the risky drinking practices of young people consuming alcohol in the night-time economy. This study will determine some of the factors and correlates associated with alcohol-related risk-taking, offending and harm in and around licensed venues and night-time entertainment precincts across five Australian cities (three metropolitan and two regional). The primary aim of the study is to measure levels of pre-drinking, drinking in venues, intoxication, illicit drug use and potentially harmful drinking practices (such as mixing with energy drinks) of patrons in entertainment areas, and relating this to offending, risky behaviour and harms experienced. The study will also investigate the effects of license type, trading hours, duration of drinking episodes and geographical location on intoxication, offending, risk-taking and experience of harm. Data collection involves patron interviews (incorporating breathalysing and drug testing) with 7500 people attending licensed venues. Intensive venue observations (n=112) will also be undertaken in a range of venues, including pubs, bars and nightclubs. The information gathered through this study will inform prevention and enforcement approaches of policy makers, police and venue staff.

Correlates of change in adults’ television viewing time : a four-year follow-up study

Purpose: Adults tend to increase their television (TV) viewing time as they age, but little is known about attributes associated with change in TV viewing over time. This study examined individual, social, and environmental correlates of change in TV viewing time for 4 yr.

Methods: Adult participants (n = 897) from a longitudinal epidemiological study in Adelaide, Australia, reported TV viewing time at baseline (2003–2004) and at follow- up (2007–2008). Generalized linear modeling was used to examine correlates of change in TV viewing time.

Results: The mean TV viewing time increased from 112 to 116 min·d-1 from baseline to follow-up. Adjusted for TV viewing time at baseline, having a tertiary education was associated with a 13% lower TV time at follow-up (P = 0.007). Each additional hour of occupational and transport physical activity at baseline was associated with a 2% and 7% lower TV viewing at follow-up (P = 0.031 and P = 0.023, respectively). For men, an additional hour of domestic physical activity was associated with a 7% higher TV viewing time at follow-up (P = 0.006). A significant neighborhood walkability &times; working status interaction (P = 0.035) indicated that, for those who were not working, living in a highly walkable neighborhood was associated with a 23% lower TV viewing time at follow-up (P = 0.003).

Conclusions: Adults with lower educational attainment, adults with lower occupational and transport physical activity, men with higher domestic physical activity, and nonworking adults living in lowly walkable neighborhoods were at higher risk of increase in TV viewing time. Interventions should target multiple variables at the individual, social, and environmental levels to address age-related increases in TV viewing time.

Independent and joint associations of TV viewing time and snack food consumption with the metabolic syndrome and its components; a cross-sectional study in Australian adults

Background:
Television (TV) viewing time is positively associated with the metabolic syndrome (MetS) in adults. However, the mechanisms through which TV viewing time is associated with MetS risk remain unclear. There is evidence that the consumption of energy-dense, nutrient poor snack foods increases during TV viewing time among adults, suggesting that these behaviors may jointly contribute towards MetS risk. While the association between TV viewing time and the MetS has previously been shown to be independent of adult’s overall dietary intake, the specific influence of snack food consumption on the relationship is yet to be investigated. The purpose of this study was to examine the independent and joint associations of daily TV viewing time and snack food consumption with the MetS and its components in a sample of Australian adults.

Methods:
Population-based, cross-sectional study of 3,110 women and 2,572 men (>35 years) without diabetes or cardiovascular disease. Participants were recruited between May 1999 and Dec 2000 in the six states and the Northern Territory of Australia. Participants were categorised according to self-reported TV viewing time (low: 0-2 hr/d; high: >2 hr/d) and/or consumption of snack foods (low: 0-3 serves/d; high: >3 serves/d). Multivariate odds ratios [95% CI] for the MetS and its components were estimated using gender-specific, forced entry logistic regression.

Results:
OR [95% CI] for the MetS was 3.59 [2.25, 5.74] (p≤0.001) in women and 1.45 [1.02, 3.45] (p = 0.04) in men who jointly reported high TV viewing time and high snack food consumption. Obesity, insulin resistance and hypertension (women only) were also jointly associated with high TV viewing time and high snack food consumption. Further adjustment for diet quality and central adiposity maintained the associations in women. High snack food consumption was also shown to be independently associated with MetS risk [OR: 1.94 (95% CI: 1.45, 2.60), p < 0.001] and hypertension [OR: 1.43 (95% CI: 1.01, 2.02), p = 0.05] in women only. For both men and women, high TV viewing time was independently associated with the MetS and its individual components (except hypertension).

Conclusion:
TV viewing time and snack food consumption are independently and jointly associated with the MetS and its components, particularly in women. In addition to physical activity, population strategies targeting MetS prevention should address high TV time and excessive snack food intake.

Reducing office workers' sitting time: rationale and study design for the Stand Up Victoria cluster randomized trial

Background
Excessive time spent in sedentary behaviours (sitting or lying with low energy expenditure) is associated with an increased risk for type 2 diabetes, cardiovascular disease and some cancers. Desk-based office workers typically accumulate high amounts of daily sitting time, often in prolonged unbroken bouts. The Stand Up Victoria study aims to determine whether a 3-month multi-component intervention in the office setting reduces workplace sitting, particularly prolonged, unbroken sitting time, and results in improvements in cardiometabolic biomarkers and work-related outcomes, compared to usual practice.

Methods/design
A two-arm cluster-randomized controlled trial (RCT), with worksites as the unit of randomization, will be conducted in 16 worksites located in Victoria, Australia. Work units from one organisation (Department of Human Services, Australian Government) will be allocated to either the multi-component intervention (organisational, environmental [heightadjustable workstations], and individual behavioural strategies) or to a usual practice control group. The recruitment target is 160 participants (office-based workers aged 18–65 years and working at least 0.6 full time equivalent) per arm. At each assessment (0- [baseline], 3- [post intervention], and 12-months [follow-up]), objective measurement via the activPAL3 activity monitor will be used to assess workplace: sitting time (primary outcome); prolonged sitting time (sitting time accrued in bouts of ≥30 minutes); standing time; sit-to-stand transitions; and, moving time. Additional outcomes assessed will include: non-workplace activity; cardio-metabolic biomarkers and health indicators (including fasting glucose, lipids and insulin; anthropometric measures; blood pressure; and, musculoskeletal symptoms); and, work-related outcomes (presenteeism, absenteeism, productivity, work performance). Incremental cost-effectiveness and identification of both workplace and individual-level mediators and moderators of change will also be evaluated.

Dynamic association between interpersonal functioning and positive symptom dimensions of psychosis over time: a longitudinal study of healthy adolescents

Cross-sectional studies have indicated that alterations in social functioning, particularly interpersonal functioning, are associated with the occurrence of psychotic symptoms and experiences at different levels of the extended psychosis phenotype (ranging from population psychometric expression of liability to overt psychotic disorder). However, more research is needed on the development of this association over time

International study of objectively measured physical activity and sedentary time with body mass index and obesity: IPEN adult study

BACKGROUND: Physical activity (PA) has been consistently implicated in the etiology of obesity, whereas recent evidence on the importance of sedentary time remains inconsistent. Understanding of dose-response associations of PA and sedentary time with overweight and obesity in adults can be improved with large-scale studies using objective measures of PA and sedentary time. The purpose of this study was to examine the strength, direction and shape of dose-response associations of accelerometer-based PA and sedentary time with body mass index (BMI) and weight status in 10 countries, and the moderating effects of study site and gender. METHODS: Data from the International Physical activity and the Environment Network (IPEN) Adult study were used. IPEN Adult is an observational multi-country cross-sectional study, and 12 sites in 10 countries are included. Participants wore an accelerometer for seven consecutive days, completed a socio-demographic questionnaire and reported height and weight. In total, 5712 adults (18-65 years) were included in the analyses. Generalized additive mixed models, conducted in R, were used to estimate the strength and shape of the associations. RESULTS: A curvilinear relationship of accelerometer-based moderate-to-vigorous PA and total counts per minute with BMI and the probability of being overweight/obese was identified. The associations were negative, but weakened at higher levels of moderate-to-vigorous PA (>50 min per day) and higher counts per minute. No associations between sedentary time and weight outcomes were found. Complex site- and gender-specific findings were revealed for BMI, but not for weight status. CONCLUSIONS: On the basis of these results, the current Institute of Medicine recommendation of 60 min per day of moderate-to-vigorous PA to prevent weight gain in normal-weight adults was supported. No relationship between sedentary time and the weight outcomes was present, calling for further examination. If moderator findings are confirmed, the relationship between PA and BMI may be country- and gender-dependent, which could have important implications for country-specific health guidelines. © 2015 Macmillan Publishers Limited All rights reserved.