Distributed classification for pocket data mining

Distributed and collaborative data stream mining in a mobile computing environment is referred to as Pocket Data Mining PDM. Large amounts of available data streams to which smart phones can subscribe to or sense, coupled with the increasing computational power of handheld devices motivates the development of PDM as a decision making system. This emerging area of study has shown to be feasible in an earlier study using technological enablers of mobile software agents and stream mining techniques [1]. A typical PDM process would start by having mobile agents roam the network to discover relevant data streams and resources. Then other (mobile) agents encapsulating stream mining techniques visit the relevant nodes in the network in order to build evolving data mining models. Finally, a third type of mobile agents roam the network consulting the mining agents for a final collaborative decision, when required by one or more users. In this paper, we propose the use of distributed Hoeffding trees and Naive Bayes classifers in the PDM framework over vertically partitioned data streams. Mobile policing, health monitoring and stock market analysis are among the possible applications of PDM. An extensive experimental study is reported showing the effectiveness of the collaborative data mining with the two classifers.

Homogeneous and heterogeneous distributed classification for pocket data mining

Pocket Data Mining (PDM) describes the full process of analysing data streams in mobile ad hoc distributed environments. Advances in mobile devices like smart phones and tablet computers have made it possible for a wide range of applications to run in such an environment. In this paper, we propose the adoption of data stream classification techniques for PDM. Evident by a thorough experimental study, it has been proved that running heterogeneous/different, or homogeneous/similar data stream classification techniques over vertically partitioned data (data partitioned according to the feature space) results in comparable performance to batch and centralised learning techniques.

P-found: grid-enabling distributed repositories of protein folding and unfolding simulations for data mining

The P-found protein folding and unfolding simulation repository is designed to allow scientists to perform data mining and other analyses across large, distributed simulation data sets. There are two storage components in P-found: a primary repository of simulation data that is used to populate the second component, and a data warehouse that contains important molecular properties. These properties may be used for data mining studies. Here we demonstrate how grid technologies can support multiple, distributed P-found installations. In particular, we look at two aspects: firstly, how grid data management technologies can be used to access the distributed data warehouses; and secondly, how the grid can be used to transfer analysis programs to the primary repositories — this is an important and challenging aspect of P-found, due to the large data volumes involved and the desire of scientists to maintain control of their own data. The grid technologies we are developing with the P-found system will allow new large data sets of protein folding simulations to be accessed and analysed in novel ways, with significant potential for enabling scientific discovery.

Demand response — a different form of distributed storage?

Reduced flexibility of low carbon generation could pose new challenges for future energy systems. Both demand response and distributed storage may have a role to play in supporting future system balancing. This paper reviews how these technically different, but functionally similar approaches compare and compete with one another. Household survey data is used to test the effectiveness of price signals to deliver demand responses for appliances with a high degree of agency. The underlying unit of storage for different demand response options is discussed, with particular focus on the ability to enhance demand side flexibility in the residential sector. We conclude that a broad range of options, with different modes of storage, may need to be considered, if residential demand flexibility is to be maximised.

Fertilisers and insect herbivores: a meta-analysis

Despite the importance of a thorough understanding of the effect of synthetic fertiliser on insect population dynamics, existing literature is conflicting and an area of intense debate. Here, a categorical random-effects meta-analysis and a vote count meta-analysis are employed to examine the effects of nitrogen(N), phosphorus (P), potassium (K) and NPK fertiliser on insect population dynamics. In agreement with the general consensus, insects were found to respond positively, overall, to fertilisers. Sucking insects showed a much stronger response to fertilisers than chewing insects. The environment in which a study is conducted can have a marked effect on insect responses to fertiliser, with natural environments showing the potential for buffering effects of nitrogen fertilisers in particular. As well as highlighting the potential shortfall in the amount of research investigating particularly the effects of potassium and phosphorus, this study provides an invaluable flag post in the ongoing research investigating fertiliser effects on ecosystems.

Scaling up data mining techniques to large datasets using parallel and distributed processing

Advances in hardware and software technology enable us to collect, store and distribute large quantities of data on a very large scale. Automatically discovering and extracting hidden knowledge in the form of patterns from these large data volumes is known as data mining. Data mining technology is not only a part of business intelligence, but is also used in many other application areas such as research, marketing and financial analytics. For example medical scientists can use patterns extracted from historic patient data in order to determine if a new patient is likely to respond positively to a particular treatment or not; marketing analysts can use extracted patterns from customer data for future advertisement campaigns; finance experts have an interest in patterns that forecast the development of certain stock market shares for investment recommendations. However, extracting knowledge in the form of patterns from massive data volumes imposes a number of computational challenges in terms of processing time, memory, bandwidth and power consumption. These challenges have led to the development of parallel and distributed data analysis approaches and the utilisation of Grid and Cloud computing. This chapter gives an overview of parallel and distributed computing approaches and how they can be used to scale up data mining to large datasets.

Power allocation strategies for distributed space-time codes in two-way relay networks

We study a two-way relay network (TWRN), where distributed space-time codes are constructed across multiple relay terminals in an amplify-and-forward mode. Each relay transmits a scaled linear combination of its received symbols and their conjugates,with the scaling factor chosen based on automatic gain control. We consider equal power allocation (EPA) across the relays, as well as the optimal power allocation (OPA) strategy given access to instantaneous channel state information (CSI). For EPA, we derive an upper bound on the pairwise-error-probability (PEP), from which we prove that full diversity is achieved in TWRNs. This result is in contrast to one-way relay networks, in which case a maximum diversity order of only unity can be obtained. When instantaneous CSI is available at the relays, we show that the OPA which minimizes the conditional PEP of the worse link can be cast as a generalized linear fractional program, which can be solved efficiently using the Dinkelback-type procedure.We also prove that, if the sum-power of the relay terminals is constrained, then the OPA will activate at most two relays.

Use of nutritional complete supplements in older adults with dementia: systematic review and meta-analysis of clinical outcomes

Background & Aims: Malnutrition is prevalent in people diagnosed with dementia however ensuring adequate oral intake within this group is often problematic. It is important to determine whether providing nutritionally complete oral nutritional supplements (ONS) drinks is an effective way of improving clinical outcomes for older people with dementia. This paper systematically reviewed clinical, wellbeing and nutritional outcomes in people with long-term cognitive impairment. Methods: The CINAHL, Medline and EMBASE databases were searched from their inception until January 2012. Reference lists of the included papers, foreign language papers and review articles obtained were manually searched. Results: Twelve articles were included in the review containing 1076 people in the supplement groups (intervention) and 748 people in the control groups. Meta-analysis shows there was a significant improvement in weight (p=<0.0001), Body Mass Index (BMI) (p=<0.0001) and cognition at 6.5+/-3.9 month follow up (p=0.002) when supplements were given compared to the control group. Conclusions: Providing ONS drinks has a positive effect on weight gain and cognition at follow up in older people with dementia. Additional research is required in both comparing nutritional supplements to vitamin/mineral tablets and high protein/calorie shots and clinical outcomes relevant to hospitalised people with dementia.

Environmental factors driving the effectiveness of European agri-environmental measures in mitigating pollinator loss – a meta-analysis

In Europe, agri-environmental schemes (AES) have been introduced in response to concerns about farmland biodiversity declines. Yet, as AES have delivered variable results, a better understanding of what determines their success or failure is urgently needed. Focusing on pollinating insects, we quantitatively reviewed how environmental factors affect the effectiveness of AES. Our results suggest that the ecological contrast in floral resources created by schemes drives the response of pollinators to AES but that this response is moderated by landscape context and farmland type, with more positive responses in croplands (vs. grasslands) located in simple (vs. cleared or complex) landscapes. These findings inform us how to promote pollinators and associated pollination services in species-poor landscapes. They do not, however, present viable strategies to mitigate loss of threatened or endangered species. This indicates that the objectives and design of AES should distinguish more clearly between biodiversity conservation and delivery of ecosystem services.

Comparison of pollinators and natural enemies: a meta-analysis of landscape and local effects on abundance and richness in crops

To manage agroecosystems for multiple ecosystem services, we need to know whether the management of one service has positive, negative, or no effects on other services. We do not yet have data on the interactions between pollination and pest-control services. However, we do have data on the distributions of pollinators and natural enemies in agroecosystems. Therefore, we compared these two groups of ecosystem service providers, to see if the management of farms and agricultural landscapes might have similar effects on the abundance and richness of both. In a meta-analysis, we compared 46 studies that sampled bees, predatory beetles, parasitic wasps, and spiders in fields, orchards, or vineyards of food crops. These studies used the proximity or proportion of non-crop or natural habitats in the landscapes surrounding these crops (a measure of landscape complexity), or the proximity or diversity of non-crop plants in the margins of these crops (a measure of local complexity), to explain the abundance or richness of these beneficial arthropods. Compositional complexity at both landscape and local scales had positive effects on both pollinators and natural enemies, but different effects on different taxa. Effects on bees and spiders were significantly positive, but effects on parasitoids and predatory beetles (mostly Carabidae and Staphylinidae) were inconclusive. Landscape complexity had significantly stronger effects on bees than it did on predatory beetles and significantly stronger effects in non-woody rather than in woody crops. Effects on richness were significantly stronger than effects on abundance, but possibly only for spiders. This abundance-richness difference might be caused by differences between generalists and specialists, or between arthropods that depend on non-crop habitats (ecotone species and dispersers) and those that do not (cultural species). We call this the ‘specialist-generalist’ or ‘cultural difference’ mechanism. If complexity has stronger effects on richness than abundance, it might have stronger effects on the stability than the magnitude of these arthropod-mediated ecosystem services. We conclude that some pollinators and natural enemies seem to have compatible responses to complexity, and it might be possible to manage agroecosystems for the benefit of both. However, too few studies have compared the two, and so we cannot yet conclude that there are no negative interactions between pollinators and natural enemies, and no trade-offs between pollination and pest-control services. Therefore, we suggest a framework for future research to bridge these gaps in our knowledge.

The competition–performance relation: a meta-analytic review and test of the opposing processes model of competition and performance

What is the relation between competition and performance? The present research addresses this important multidisciplinary question by conducting a meta-analysis of existing empirical work and by proposing a new conceptual model—the opposing processes model of competition and performance. This model was tested by conducting an additional meta-analysis and 3 new empirical studies. The first meta-analysis revealed that there is no noteworthy relation between competition and performance. The second meta-analysis showed, in accord with the opposing processes model, that the absence of a direct effect is the result of inconsistent mediation via achievement goals: Competition prompts performance-approach goals which, in turn, facilitate performance; and competition also prompts performance-avoidance goals which, in turn, undermine performance. These same direct and mediational findings were also observed in the 3 new empirical studies (using 3 different conceptualizations of competition and attending to numerous control variables). Our findings provide both interpretational clarity regarding past research and conceptual guidance regarding future research on the competition–performance relation.

A genome-wide association meta-analysis identifies new childhood obesity loci

Multiple genetic variants have been associated with adult obesity and a few with severe obesity in childhood; however, less progress has been made in establishing genetic influences on common early-onset obesity. We performed a North American, Australian and European collaborative meta-analysis of 14 studies consisting of 5,530 cases (≥95th percentile of body mass index (BMI)) and 8,318 controls (<50th percentile of BMI) of European ancestry. Taking forward the eight newly discovered signals yielding association with P < 5 × 10(-6) in nine independent data sets (2,818 cases and 4,083 controls), we observed two loci that yielded genome-wide significant combined P values near OLFM4 at 13q14 (rs9568856; P = 1.82 × 10(-9); odds ratio (OR) = 1.22) and within HOXB5 at 17q21 (rs9299; P = 3.54 × 10(-9); OR = 1.14). Both loci continued to show association when two extreme childhood obesity cohorts were included (2,214 cases and 2,674 controls). These two loci also yielded directionally consistent associations in a previous meta-analysis of adult BMI(1).

Meta-analysis of the INSIG2 association with obesity including 74,345 individuals: does heterogeneity of estimates relate to study design?

The INSIG2 rs7566605 polymorphism was identified for obesity (BMI> or =30 kg/m(2)) in one of the first genome-wide association studies, but replications were inconsistent. We collected statistics from 34 studies (n = 74,345), including general population (GP) studies, population-based studies with subjects selected for conditions related to a better health status ('healthy population', HP), and obesity studies (OB). We tested five hypotheses to explore potential sources of heterogeneity. The meta-analysis of 27 studies on Caucasian adults (n = 66,213) combining the different study designs did not support overall association of the CC-genotype with obesity, yielding an odds ratio (OR) of 1.05 (p-value = 0.27). The I(2) measure of 41% (p-value = 0.015) indicated between-study heterogeneity. Restricting to GP studies resulted in a declined I(2) measure of 11% (p-value = 0.33) and an OR of 1.10 (p-value = 0.015). Regarding the five hypotheses, our data showed (a) some difference between GP and HP studies (p-value = 0.012) and (b) an association in extreme comparisons (BMI> or =32.5, 35.0, 37.5, 40.0 kg/m(2) versus BMI<25 kg/m(2)) yielding ORs of 1.16, 1.18, 1.22, or 1.27 (p-values 0.001 to 0.003), which was also underscored by significantly increased CC-genotype frequencies across BMI categories (10.4% to 12.5%, p-value for trend = 0.0002). We did not find evidence for differential ORs (c) among studies with higher than average obesity prevalence compared to lower, (d) among studies with BMI assessment after the year 2000 compared to those before, or (e) among studies from older populations compared to younger. Analysis of non-Caucasian adults (n = 4889) or children (n = 3243) yielded ORs of 1.01 (p-value = 0.94) or 1.15 (p-value = 0.22), respectively. There was no evidence for overall association of the rs7566605 polymorphism with obesity. Our data suggested an association with extreme degrees of obesity, and consequently heterogeneous effects from different study designs may mask an underlying association when unaccounted for. The importance of study design might be under-recognized in gene discovery and association replication so far.

The Gly482Ser genotype at the PPARGC1A gene and elevated blood pressure: a meta-analysis involving 13,949 individuals.

The protein encoded by the PPARGC1A gene is expressed at high levels in metabolically active tissues and is involved in the control of oxidative stress via reactive oxygen species detoxification. Several recent reports suggest that the PPARGC1A Gly482Ser (rs8192678) missense polymorphism may relate inversely with blood pressure. We used conventional meta-analysis methods to assess the association between Gly482Ser and systolic (SBP) or diastolic blood pressures (DBP) or hypertension in 13,949 individuals from 17 studies, of which 6,042 were previously unpublished observations. The studies comprised cohorts of white European, Asian, and American Indian adults, and adolescents from South America. Stratified analyses were conducted to control for population stratification. Pooled genotype frequencies were 0.47 (Gly482Gly), 0.42 (Gly482Ser), and 0.11 (Ser482Ser). We found no evidence of association between Gly482Ser and SBP [Gly482Gly: mean = 131.0 mmHg, 95% confidence interval (CI) = 130.5-131.5 mmHg; Gly482Ser mean = 133.1 mmHg, 95% CI = 132.6-133.6 mmHg; Ser482Ser: mean = 133.5 mmHg, 95% CI = 132.5-134.5 mmHg; P = 0.409] or DBP (Gly482Gly: mean = 80.3 mmHg, 95% CI = 80.0-80.6 mmHg; Gly482Ser mean = 81.5 mmHg, 95% CI = 81.2-81.8 mmHg; Ser482Ser: mean = 82.1 mmHg, 95% CI = 81.5-82.7 mmHg; P = 0.651). Contrary to previous reports, we did not observe significant effect modification by sex (SBP, P = 0.966; DBP, P = 0.715). We were also unable to confirm the previously reported association between the Ser482 allele and hypertension [odds ratio: 0.97, 95% CI = 0.87-1.08, P = 0.585]. These results were materially unchanged when analyses were focused on whites only. However, statistical evidence of gene-age interaction was apparent for DBP [Gly482Gly: 73.5 (72.8, 74.2), Gly482Ser: 77.0 (76.2, 77.8), Ser482Ser: 79.1 (77.4, 80.9), P = 4.20 x 10(-12)] and SBP [Gly482Gly: 121.4 (120.4, 122.5), Gly482Ser: 125.9 (124.6, 127.1), Ser482Ser: 129.2 (126.5, 131.9), P = 7.20 x 10(-12)] in individuals <50 yr (n = 2,511); these genetic effects were absent in those older than 50 yr (n = 5,088) (SBP, P = 0.41; DBP, P = 0.51). Our findings suggest that the PPARGC1A Ser482 allele may be associated with higher blood pressure, but this is only apparent in younger adults.

A generalised Bayesian instrumental variable approach under student t-distributed errors with application

Bayesian analysis is given of an instrumental variable model that allows for heteroscedasticity in both the structural equation and the instrument equation. Specifically, the approach for dealing with heteroscedastic errors in Geweke (1993) is extended to the Bayesian instrumental variable estimator outlined in Rossi et al. (2005). Heteroscedasticity is treated by modelling the variance for each error using a hierarchical prior that is Gamma distributed. The computation is carried out by using a Markov chain Monte Carlo sampling algorithm with an augmented draw for the heteroscedastic case. An example using real data illustrates the approach and shows that ignoring heteroscedasticity in the instrument equation when it exists may lead to biased estimates.