A generic classifier-ensemble approach for biomedical named entity recognition

In named entity recognition (NER) for biomedical literature, approaches based on combined classifiers have demonstrated great performance improvement compared to a single (best) classifier. This is mainly owed to sufficient level of diversity exhibited among classifiers, which is a selective property of classifier set. Given a large number of classifiers, how to select different classifiers to put into a classifier-ensemble is a crucial issue of multiple classifier-ensemble design. With this observation in mind, we proposed a generic genetic classifier-ensemble method for the classifier selection in biomedical NER. Various diversity measures and majority voting are considered, and disjoint feature subsets are selected to construct individual classifiers. A basic type of individual classifier – Support Vector Machine (SVM) classifier is adopted as SVM-classifier committee. A multi-objective Genetic algorithm (GA) is employed as the classifier selector to facilitate the ensemble classifier to improve the overall sample classification accuracy. The proposed approach is tested on the benchmark dataset – GENIA version 3.02 corpus, and compared with both individual best SVM classifier and SVM-classifier ensemble algorithm as well as other machine learning methods such as CRF, HMM and MEMM. The results show that the proposed approach outperforms other classification algorithms and can be a useful method for the biomedical NER problem.

School and individual-level characteristics are associated with children's moderate to vigorous intensity physical activity during school recess

Objective: The objective of this study was to identify school environmental characteristics associated with moderate to vigorous physical activity during school recess, including morning and lunch breaks.

Methods: Accelerometry data, child-level characteristics and school physical activity, policy and socio-cultural data were collected from 408 sixth grade children (mean age 11 years) attending 27 metropolitan primary schools in Perth, Western Australia. Hierarchical modelling identified key characteristics associated with children's recess moderate to vigorous physical activity (RMVPA).

Results: Nearly 40% of variability in children's RMVPA was explained by school environment and individual characteristics identified in this study. Children's higher daily RMVPA was associated with newer schools, schools with a higher number of grassed surfaces per child and fewer shaded grassed surfaces, and the physical education coordinator meeting Australian physical activity guidelines.

Conclusions: Characteristics of the school physical and social environments are strongly correlated with children's MPVA during recess.

Comparative evaluation of multi-label classification methods

This paper presents a comparative evaluation of popular multi-label classification methods on several multi-label problems from different domains. The methods include multi-label k-nearest neighbor, binary relevance, label power set, random k-label set ensemble learning, calibrated label ranking, hierarchy of multi-label classifiers and triple random ensemble multi-label classification algorithms. These multi-label learning algorithms are evaluated using several widely used MLC evaluation metrics. The evaluation results show that for each multi-label classification problem a particular MLC method can be recommended. The multi-label evaluation datasets used in this study are related to scene images, multimedia video frames, diagnostic medical report, email messages, emotional music data, biological genes and multi-structural proteins categorization.

Improving classifications for cardiac autonomic neuropathy using multi-level ensemble classifiers and feature selection based on random forest

This paper is devoted to empirical investigation of novel multi-level ensemble meta classifiers for the detection and monitoring of progression of cardiac autonomic neuropathy, CAN, in diabetes patients. Our experiments relied on an extensive database and concentrated on ensembles of ensembles, or multi-level meta classifiers, for the classification of cardiac autonomic neuropathy progression. First, we carried out a thorough investigation comparing the performance of various base classifiers for several known sets of the most essential features in this database and determined that Random Forest significantly and consistently outperforms all other base classifiers in this new application. Second, we used feature selection and ranking implemented in Random Forest. It was able to identify a new set of features, which has turned out better than all other sets considered for this large and well-known database previously. Random Forest remained the very best classier for the new set of features too. Third, we investigated meta classifiers and new multi-level meta classifiers based on Random Forest, which have improved its performance. The results obtained show that novel multi-level meta classifiers achieved further improvement and obtained new outcomes that are significantly better compared with the outcomes published in the literature previously for cardiac autonomic neuropathy.

Regulation to create environments conducive to physical activity : understanding the barriers and facilitators at the Australian State Government level

Introduction Policy and regulatory interventions aimed at creating environments more conducive to physical activity (PA) are an important component of strategies to improve population levels of PA. However, many potentially effective policies are not being broadly implemented. This study sought to identify potential policy/regulatory interventions targeting PA environments, and barriers/facilitators to their implementation at the Australian state/territory government level.

Methods In-depth interviews were conducted with senior representatives from state/territory governments, statutory authorities and non-government organisations (n = 40) to examine participants': 1) suggestions for regulatory interventions to create environments more conducive to PA; 2) support for preselected regulatory interventions derived from a literature review. Thematic and constant comparative analyses were conducted.

Results Policy interventions most commonly suggested by participants fell into two areas: 1) urban planning and provision of infrastructure to promote active travel; 2) discouraging the use of private motorised vehicles. Of the eleven preselected interventions presented to participants, interventions relating to walkability/cycling and PA facilities received greatest support. Interventions involving subsidisation (of public transport, PA-equipment) and the provision of more public transport infrastructure received least support. These were perceived as not economically viable or unlikely to increase PA levels. Dominant barriers were: the powerful ‘road lobby’, weaknesses in the planning system and the cost of potential interventions. Facilitators were: the provision of evidence, collaboration across sectors, and synergies with climate change/environment agendas.

Conclusion This study points to how difficult it will be to achieve policy change when there is a powerful ‘road lobby’ and government investment prioritises road infrastructure over PA-promoting infrastructure. It highlights the pivotal role of the planning and transport sectors in implementing PA-promoting policy, however suggests the need for clearer guidelines and responsibilities for state and local government levels in these areas. Health outcomes need to be given more direct consideration and greater priority within non-health sectors.

EPODE approach for childhood obesity prevention : methods, progress and international development

Childhood obesity is a complex issue and needs multistakeholder involvement at all levels to foster healthier lifestyles in a sustainable way. ‘Ensemble Prévenons l'ObésitéDes Enfants’ (EPODE, Together Let's Prevent Childhood Obesity) is a large-scale, coordinated, capacity-building approach for communities to implement effective and sustainable strategies to prevent childhood obesity. This paper describes EPODE methodology and its objective of preventing childhood obesity.

At a central level, a coordination team, using social marketing and organizational techniques, trains and coaches a local project manager nominated in each EPODE community by the local authorities. The local project manager is also provided with tools to mobilize local stakeholders through a local steering committee and local networks. The added value of the methodology is to mobilize stakeholders at all levels across the public and the private sectors. Its critical components include political commitment, sustainable resources, support services and a strong scientific input – drawing on the evidence-base – together with evaluation of the programme.

Since 2004, EPODE methodology has been implemented in more than 500 communities in six countries. Community-based interventions are integral to childhood obesity prevention. EPODE provides a valuable model to address this challenge.