The social marketing theory-based (SMT) approach for designing interventions

- Purpose Despite the importance of theory as a driving framework, many social marketers either fail to explicitly use theory as the basis of designing social marketing interventions or default to familiar theories which may not accurately reflect the nature of the behavioural issue. The purpose of this paper is therefore to propose and demonstrate the social marketing theory (SMT)-based approach for designing social marketing interventions, campaigns or tools. - Design/methodology/approach This conceptual paper proposes a four-step process and illustrates this process by applying the SMT-based approach to the digital component of a social marketing intervention for preventing domestic violence. - Findings For effective social marketing interventions, the underpinning theory must reflect consumer insights and key behavioural drivers and be used explicitly in the design process. - Practical implications Social marketing practitioners do not always understand how to use theory in the design of interventions, campaigns or tools, and scholars do not always understand how to translate theories into practice. This paper outlines a process and illustrates how theory can be selected and applied. - Originality/value This paper proposes a process for theory selection and use in a social marketing context.

EUSMT: incorporating linguistic information to SMT for a morphologically rich language. Its use in SMT-RBMT-EBMT hybridation

148 p.: graf.

改进的以SMT为基础的实时系统限界模型检测

基于SAT的限界模型检测在处理实时系统时具有很高的复杂度.SMT求解器在计算可满足性的同时,还能处理算术和其他可判定性理论.在对实时系统进行检测时,用SMT求解器代替SAT求解器,系统里的时钟就可以用整型或实型变量表示,时钟约束则可以直接表示成线性算术表达式,从而使整个检测过程更加高效.带时间参数的计算树逻辑(timed computation tree logic,简称TCTL)被用来描述实时系统里的性质.同时,还对检测方法作了相应的改进.

Managing SMT Resource Usage through Speculative Instruction Window Weighting

Simultaneous multithreading processors dynamically share processor resources between multiple threads. In general, shared SMT resources may be managed explicitly, for instance, by dynamically setting queue occupation bounds for each thread as in the DCRA and Hill-Climbing policies. Alternatively, resources may be managed implicitly; that is, resource usage is controlled by placing the desired instruction mix in the resources. In this case, the main resource management tool is the instruction fetch policy which must predict the behavior of each thread (branch mispredictions, long-latency loads, etc.) as it fetches instructions.

Turbo-smt: Accelerating coupled sparse matrix-tensor factorizations by 200x

How can we correlate the neural activity in the human brain as it responds to typed words, with properties of these terms (like ‘edible’, ‘fits in hand’)? In short, we want to find latent variables, that jointly explain both the brain activity, as well as the behavioral responses. This is one of many settings of the Coupled Matrix-Tensor Factorization (CMTF) problem.

Can we accelerate any CMTF solver, so that it runs within a few minutes instead of tens of hours to a day, while maintaining good accuracy? We introduce Turbo-SMT, a meta-method capable of doing exactly that: it boosts the performance of any CMTF algorithm, by up to 200x, along with an up to 65 fold increase in sparsity, with comparable accuracy to the baseline.

We apply Turbo-SMT to BrainQ, a dataset consisting of a (nouns, brain voxels, human subjects) tensor and a (nouns, properties) matrix, with coupling along the nouns dimension. Turbo-SMT is able to find meaningful latent variables, as well as to predict brain activity with competitive accuracy.

Scoup-SMT: Scalable Coupled Sparse Matrix-Tensor Factorization

How can we correlate neural activity in the human brain as it responds to words, with behavioral data expressed as answers to questions about these same words? In short, we want to find latent variables, that explain both the brain activity, as well as the behavioral responses. We show that this is an instance of the Coupled Matrix-Tensor Factorization (CMTF) problem. We propose Scoup-SMT, a novel, fast, and parallel algorithm that solves the CMTF problem and produces a sparse latent low-rank subspace of the data. In our experiments, we find that Scoup-SMT is 50-100 times faster than a state-of-the-art algorithm for CMTF, along with a 5 fold increase in sparsity. Moreover, we extend Scoup-SMT to handle missing data without degradation of performance. We apply Scoup-SMT to BrainQ, a dataset consisting of a (nouns, brain voxels, human subjects) tensor and a (nouns, properties) matrix, with coupling along the nouns dimension. Scoup-SMT is able to find meaningful latent variables, as well as to predict brain activity with competitive accuracy. Finally, we demonstrate the generality of Scoup-SMT, by applying it on a Facebook dataset (users, friends, wall-postings); there, Scoup-SMT spots spammer-like anomalies.

Alignment Model and Training Technique in SMT from English to Malayalam

This paper investigates certain methods of training adopted in the Statistical Machine Translator (SMT) from English to Malayalam. In English Malayalam SMT, the word to word translation is determined by training the parallel corpus. Our primary goal is to improve the alignment model by reducing the number of possible alignments of all sentence pairs present in the bilingual corpus. Incorporating morphological information into the parallel corpus with the help of the parts of speech tagger has brought around better training results with improved accuracy

Arquiteturas multi-tarefas simultâneas : SEMPRE : arquitetura SMT com capacidade de execução e escalonamento de processos

O avanço tecnológico no projeto de microprocessadores, nos recentes anos, tem seguido duas tendências principais. A primeira tenta aumentar a freqüência do relógio dos mesmos usando componentes digitais e técnicas VLSI mais eficientes. A segunda tenta explorar paralelismo no nível de instrução através da reorganização dos seus componentes internos. Dentro desta segunda abordagem estão as arquiteturas multi-tarefas simultâneas, que são capazes de extrair o paralelismo existente entre e dentro de diferentes tarefas das aplicações, executando instruções de vários fluxos simultaneamente e maximizando assim a utilização do hardware. Apesar do alto custo da implementação em hardware, acredita-se no potencial destas arquiteturas para o futuro próximo, pois é previsto que em breve haverá a disponibilidade de bilhões de transistores para o desenvolvimento de circuitos integrados. Assim, a questão principal a ser encarada talvez seja: como prover instruções paralelas para uma arquitetura deste tipo? Sabe-se que a maioria das aplicações é seqüencial pois os problemas nem sempre possuem uma solução paralela e quando a solução existe os programadores nem sempre têm habilidade para ver a solução paralela. Pensando nestas questões a arquitetura SEMPRE foi projetada. Esta arquitetura executa múltiplos processos, ao invés de múltiplas tarefas, aproveitando assim o paralelismo existente entre diferentes aplicações. Este paralelismo é mais expressivo do que aquele que existe entre tarefas dentro de uma mesma aplicação devido a não existência de sincronismo ou comunicação entre elas. Portanto, a arquitetura SEMPRE aproveita a grande quantidade de processos existentes nas estações de trabalho compartilhadas e servidores de rede. Além disso, esta arquitetura provê suporte de hardware para o escalonamento de processos e instruções especiais para o sistema operacional gerenciar processos com mínimo esforço. Assim, os tempos perdidos com o escalonamento de processos e as trocas de contextos são insignificantes nesta arquitetura, provendo ainda maior desempenho durante a execução das aplicações. Outra característica inovadora desta arquitetura é a existência de um mecanismo de prébusca de processos que, trabalhando em cooperação com o escalonamento de processos, permite reduzir faltas na cache de instruções. Também, devido a essa rápida troca de contexto, a arquitetura permite a definição de uma fatia de tempo (fatia de tempo) menor do que aquela praticada pelo sistema operacional, provendo maior dinâmica na execução das aplicações. A arquitetura SEMPRE foi analisada e avaliada usando modelagem analítica e simulação dirigida por execução de programas do SPEC95. A modelagem mostrou que o escalonamento por hardware reduz os efeitos colaterais causados pela presença de processos na cache de instruções e a simulação comprovou que as diferentes características desta arquitetura podem, juntas, prover ganho de desempenho razoável sobre outras arquiteturas multi-tarefas simultâneas equivalentes, com um pequeno acréscimo de hardware, melhor aproveitando as fatias de tempo atribuídas aos processos.

Interlocked projects in safety competency and safety effectiveness indicators in the construction sector

Safety culture is a concept that has long been accepted in high risk industries such as aviation, nuclear industries and mining, however, considerable research is now also being undertaken within the construction sector. This paper discusses three recent interlocked projects undertaken in the Australian construction industry. The first project examined the development and implementation of a safety competency framework targeted at safety critical positions (SCP's) across first tier construction organisations. Combining qualitative and quantitative methods, the project: developed a matrix of SCP's (n=11) and safety management tasks (SMTs; n=39); mapped the process steps for their acquisition and development; detailed the knowledge, skills and behaviours required for all SMTs; and outlined potential organisational cultural outcomes from a successful implementation of the framework. The second project extended this research to develop behavioural guidelines for leaders to drive safety culture change down to second tier companies and to assist them to customise their own competency framework and implementation guidelines to match their aspirations and resources. The third interlocked project explored the use of safety effectiveness indicators (SEIs) as an industry-relevant assessment tool for reducing risk on construction sites. With direct linkages to safety competencies and SMT's, the SEIs are the next step towards an integrated safety cultural approach to safety and extend the concept of positive performance indicators (PPIs) by providing a valid, reliable, and user friendly measurement platform. Taken together, the results of the interlocked projects suggest that industry engaged collaborative safety culture research has many potential benefits for the construction industry.