The aim of this paper is to design an algorithm based on nonsmooth optimization techniques to solve the minimum sum-of-squares clustering problems in very large data sets. First, the clustering problem is formulated as a nonsmooth optimization problem. Then the limited memory bundle method [Haarala et al., 2007] is modified and combined with an incremental approach to design a new clustering algorithm. The algorithm is evaluated using real world data sets with both the large number of attributes and the large number of data points. It is also compared with some other optimization based clustering algorithms. The numerical results demonstrate the efficiency of the proposed algorithm for clustering in very large data sets.
Objectives: To describe patterns of time use among regional and rural adolescent girls and compare identified clusters with respect to correlates of physical activity (PA) and health-related quality of life (HRQoL). Methods: Data were from Year 7-9 adolescent girls (aged 12-15 years) from 16 schools involved in a cluster-randomised trial in regional and rural Victoria, Australia (n = 494). Time use data were collected using 24-h Previous Day Physical Activity Recall (PDPAR-24) questionnaire, collapsed into 17 categories of time use. Differences between time use clusters with regard to demographics, correlates of PA and HRQoL measured using PedsQL 4.0 Generic Core Scales, were investigated. Results: Two time use clusters were identified and were associated with correlates of PA and HRQoL. Girls who spent significantly more time in teams sports, non-team sports, school classes, watching TV and sleeping had higher levels of positively aligned PA correlates (e.g. self-efficacy, perceived sports competence) and HRQoL than girls characterised with high levels of computer use and video gaming. Conclusions: These findings highlight how different activity patterns of regional and rural girls affect HRQoL and can inform future intervention strategies to improve PA levels and HRQoL. Clusters characterised by low levels of PA and high computer use and video gaming require targeted interventions to address barriers to their participation. (C) 2016 Published by Elsevier Ltd on behalf of Sports Medicine Australia.