1,720,964 research outputs found
Inferring test models from user bug reports using multi-objective search
No full textBug reports are used by software testers to identify abnormal software behaviour. In this paper, we propose a multi-objective evolutionary approach to automatically generate finite state machines (FSMs) based on bug reports written in natural language, to automatically capture incorrect software behaviour. These FSMs can then be used by testers to both exercise the reported bugs and create tests that can potentially reveal new bugs. The FSM generation is guided by a Multi-Objective Evolutionary Algorithm (MOEA) that simultaneously minimises three objectives: size of the models, number of unrealistic states (over-generalisation), and number of states not covered by the models (under-generalisation). We assess the feasibility of our approach for 10 real-world software programs by exploiting three different MOEAs (NSGA-II, NSGA-III and MOEA/D) and benchmarking them with the baseline tool KLFA. Our results show that KLFA is not practical to be used with real-world software, because it generates models that over generalise software behaviour. Among the three MOEAs, NSGA-II obtained significantly better results than the other two for all 10 programs, detecting a greater number of bugs for 90% of the programs. We also studied the differences in quality and model performance when MOEAs are guided by only two objectives rather than three during the evolution. We found that the use of under-approximation (or over-approximation) and size as objectives generates infeasible solutions. On the other hand, using as objectives over-approximation and under-approximation generates feasible solutions yet still worse than those obtained using all three objectives for 100% of the cases. The size objective acts as a diversity factor. As a consequence, an algorithm guided by all three objectives avoids local optima, controls the size of the models, and makes the results more diverse and closer to the optimal Pareto set
Uso de padrões em projeto arquitetural baseado em busca de linha de produto de software
Full text linkOrientadora : Profª. Drª. Silvia Regina VergilioCo-orientadora : Profª. Drª. Thelma Elita Colanzi LopesDissertação (mestrado) - Universidade Federal do Paraná, Setor de Ciências Exatas, Programa de Pós-Graduação em Informática. Defesa: Curitiba, 23/10/2014Inclui referênciasResumo: Padrões de projeto visam a melhorar o entendimento e o reúso de arquiteturas de software. No projeto baseado em busca eles têm sido aplicados com sucesso por meio de operadores de mutação em processos evolutivos. No contexto de Arquiteturas de Linha de Produtos (ALPs), alguns trabalhos têm aplicado padrões de projeto manualmente, mas não existem abordagens baseadas em busca que considerem o uso destes padrões. Tornar este uso possível é o objetivo deste trabalho, que introduz uma forma automática para aplicação de padrões de projeto por meio de um operador de mutação na abordagem Multi-objective Optimization Approach for PLA Design (MOA4PLA). A ideia é que esta aplicação não gere anomalias na arquitetura e garanta que estes padrões sejam aplicados em escopos realmente propícios para suas aplicações. Para isso, foi realizada uma análise de viabilidade para determinar quais padrões do catálogo GoF (Gang of Four) são aplicáveis no contexto da MOA4PLA. Um operador de mutação é proposto para ser utilizado neste contexto, de modo a aplicar estes padrões de projeto em ALPs durante o processo evolutivo utilizando métodos de verificação de escopos e aplicação de padrões. O operador de mutação foi implementado no módulo OPLA-Patterns da ferramenta OPLA-Tool, que dá suporte a abordagem MOA4PLA. Experimentos foram configurados e executados em ALPs reais para avaliar quantitativamente e qualitativamente os resultados obtidos. Os resultados mostram que a aplicação de padrões de projeto permite a obtenção de arquiteturas com melhores valores em métricas de software resultando em uma maior diversidade de soluções para que o arquiteto possa escolher qual delas mais se adequa aos seus objetivos. Portanto, aplicar padrões de projeto por meio do operador proposto contribui positivamente para o projeto de ALP.Abstract: Design patterns aim at improving the understanding and reuse of software architectures. In the search-based design they have been successfully applied by mutation operators in the evolutionary process. In the software Product Line Architecture (PLA) context, some works have manually applied design patterns, but there are no search-based approaches that take into account the use of these patterns. To make this use possible is the goal of this work, which introduces an automated way for the application of design patterns through a mutation operator in the MOA4PLA approach (Multi-objective Optimization Approach for PLA Design). The goal is to avoid the introduction of architectural anomalies and to ensure that these patterns are applied only in feasible scopes. To this end, a feasibility analysis was conducted to determine which patterns of the GoF (Gang of Four) catalog could be applied in the context of MOA4PLA. A mutation operator is proposed in order to apply feasible design patterns in PLAs during the evolutionary process. The operator uses scope verification and design patterns application methods. The mutation operator was implemented in the OPLA-Patterns module of OPLA-Tool, which supports the MOA4PLA approach. Experiments were configured and executed in real PLAs to quantitatively and qualitatively evaluate the results. The results showed that the application of design patterns allows the generation of architectures with better values of the software metrics. A greater diversity of solutions was obtained, then the architect can choose which one best fits his/her objectives. Therefore, the application of design patterns using the proposed operator contributes positively to the PLA design
Going Beyond Counting First Authors in Author Co-citation Analysis
Full text linkThe present study examines one of the fundamental aspects of author co-citation analysis (ACA) - the way co-citation
counts are defined. Co-citation counting provides the data on which all subsequent statistical analyses and mappings
are based, and we compare ACA results based on two different types of co-citation counting - the traditional type that
only counts the first one among a cited work's authors on the one hand and a non-traditional type that takes into
account the first 5 authors of a cited work on the other hand. Results indicate that the picture produced through this non-traditional author co-citation counting contains more coherent author groups and is therefore considerably clearer. However, this picture represents fewer specialties in the research field being studied than that produced through the traditional first-author co-citation counting when the same number of top-ranked authors is selected and analyzed. Reasons for these effects are discussed
Variations on the Author
Full text link“Variations on the Author” discusses two of Eduardo Coutinho’s recent films (Um Dia na Vida, from 2010, and Últimas Conversas, posthumously released in 2015) and their contribution to the general question of documentary authorship. The director’s filmography is characterized by a consistent yet self-effacing form of authorial self-inscription: Coutinho often features as an interviewer that rather than express opinions propels discourses; an interviewer that is good at listening. This mode of self-inscription characterizes him as an author who is not expressive but who is nonetheless markedly present on the screen. In Um Dia na Vida, however, Coutinho is completely absent form the image, while Últimas Conversas, on the contrary, includes a confessional prologue that moves the director from the margins to the center of his films. This article examines the ways in which these works stand out in the filmography of a director who offers new insights into the notion of cinematic authorship
Appropriate Similarity Measures for Author Cocitation Analysis
Full text linkWe provide a number of new insights into the methodological discussion about author cocitation analysis. We first argue that the use of the Pearson correlation for measuring the similarity between authors’ cocitation profiles is not very satisfactory. We then discuss what kind of similarity measures may be used as an alternative to the Pearson correlation. We consider three similarity measures in particular. One is the well-known cosine. The other two similarity measures have not been used before in the bibliometric literature. Finally, we show by means of an example that our findings have a high practical relevance.information science;Pearson correlation;cosine;similarity measure;author cocitation analysis
Dispelling the Myths Behind First-author Citation Counts
Full text linkWe conducted a full-scale evaluative citation analysis study of scholars in the XML research field to explore just how different from each other author rankings resulting from different citation counting methods actually are, and to demonstrate the capability of emerging data and tools on the Web in supporting more realistic citation counting methods. Our results contest some common arguments for the continued
use of first-author citation counts in the evaluation of scholars, such as high correlations between author rankings by first-author citation counts and other citation
counting methods, and high costs of using more realistic citation counting methods that are not well-supported by the ISI databases. It is argued that increasingly available digital full text research papers make it possible for citation analysis studies to go beyond what the ISI databases have directly supported and to employ more
sophisticated methods
Sentinel : a multi-objective hyper-heuristic approach for the automatic generation of mutant reduction strategies
Full text linkOrientadora: Dr. Silvia Regina VergilioCoorientadora: Dr. Federica SarroTese (doutorado) - Universidade Federal do Paraná, Setor de Ciências Exatas, Programa de Pós-Graduação em Informática. Defesa : Curitiba, 03/12/2018Inclui referências: p.83-90Área de concentração: Ciência da ComputaçãoResumo: O Teste de Mutação utiliza programas mutantes gerados com pequenas modificações no programa em teste. Tais modificações são produzidas por operadores de mutação que descrevem possíveis defeitos. O objetivo é matar os mutantes com casos de teste que produzam diferentes saídas para os mutantes e para o programa original. Ao final, uma medida de avaliação do teste, o escore de mutação, é dada considerando o número de mutantes mortos em relação ao número de mutantes gerados. O teste de mutação é bastante eficaz em revelar defeitos, entretanto, possui um alto custo computacional relacionado à execução dos mutantes. Diferentes estratégias existem para reduzir o custo do teste de mutação. Por exemplo, um subconjunto de mutantes pode ser selecionado aleatoriamente, ou pode-se utilizar um conjunto menor de operadores, ou ainda algoritmos de busca para selecionar os melhores mutantes ou operadores sem reduzir o escore global. Experimentos reportados na literatura mostram que nenhuma estratégia pode ser considerada a melhor em todos os contextos. A determinação das melhores estratégias e suas combinações de parâmetros só é possível com a condução de experimentos comparando estratégias em cenários específicos, o que pode elevar o custo do teste. O uso de uma estratégia inadequada pode comprometer a eficácia do teste e implica em uma redução de custo insatisfatória. Uma abordagem automática para selecionar e configurar estratégias pode auxiliar o testador nesta tarefa. Hiper-heurísticas são opções viáveis para este propósito, uma vez que estas são utilizadas para selecionar ou gerar boas heurísticas (estratégias) ao invés de tentar resolver o problema diretamente. Hiper-heurísticas vêm sendo estudadas na literatura, e bons resultados foram obtidos na área de Engenharia de Software Baseada em Busca. Todavia, poucos trabalhos investigam a aplicação de hiper-heurísticas no teste de mutação, e nenhum deles auxilia na geração ou configuração de estratégias de teste de mutação. Diante disto, este trabalho propõe uma abordagem baseada em hiper-heurísticas chamada Sentinel. Essa abordagem tem como objetivo principal gerar automaticamente estratégias de redução de mutantes de modo a reduzir o custo do teste de mutação sem comprometer a eficácia do teste em termos de escore de mutação e a habilidade de revelar defeitos. Sentinel utiliza elementos provenientes de diversos tipos de estratégias de redução de mutantes e os combina de modo a gerar novas estratégias potencialmente mais efetivas. A ideia é que o testador execute Sentinel, reutilize as estratégias geradas automaticamente e consequentemente não necessite escolher e nem configurar as estratégias manualmente. Isso facilita a aplicação do teste de mutação e reduz assim o custo da atividade de teste como um todo. Para avaliar a viabilidade de Sentinel, múltiplos experimentos foram executados. Os experimentos foram conduzidos em 10 diferentes sistemas com 4 versões cada, totalizando 40 sistemas. As estratégias geradas por Sentinel se mostraram significativamente mais efetivas do que estratégias convencionais da literatura. Além disso, elas mantiveram os seus bons resultados em novas versões dos softwares, o que demonstrou a sua reusabilidade. Palavras-chave: hiper-heurística, teste de software, geração de estratégias, redução de mutantes, evolução gramatical, engenharia de software baseada em busca, otimização multiobjetivo.Abstract: Mutation Testing uses mutant programs generated with small modifications in the program under test. Such modifications are produced by mutation operators that describe possible faults. The objective is to kill the mutants with test cases that produce different outputs for the mutants and the original program. In the end, a test assessment measure, called mutation score, is computed considering the number of dead mutants with respect to the number of generated ones. Mutation testing is very efficacious in revealing faults, however, it has a high computational cost related to the execution of mutants. There are several strategies to reduce the mutation testing cost. For instance, a subset of mutants can be randomly selected, or a smaller set of operators can be used, or even search based algorithms can be applied to select the best mutants or operators without reducing the global mutation score. Experiments reported in the literature show that no strategy has been proven to be the best one in all contexts. The determination of the best strategies and their parameter combination is only possible by conducting experiments comparing the strategies in specific scenarios, which can increase the testing cost. The use of an inadequate strategy might compromise the test efficacy and implies in unsatisfactory cost reduction. An automatic approach to select and configure strategies can assist the tester in this task. Hyper-heuristics are viable options for this end, since they are used for selecting or generating heuristics (strategies) instead of trying to solve the problem directly. Hyper-heuristics have been studied in the literature and good results were obtained in the Search Based Software Engineering field. However, few works have investigated the usage of hyper-heuristics in mutation testing, and none of them support the generation or configuration of mutation testing strategies. In light of this, this work proposes a hyper-heuristic based approach called Sentinel. This approach has as main objective the automatic generation of mutant reduction strategies for the cost reduction of mutation testing without compromising the test efficacy in terms of mutation score and ability in revealing faults. Sentinel uses features from several kinds of mutant reduction strategies and combines them to generate new strategies that are potentially more effective. The idea is that the tester executes Sentinel, reuses the automatically generated strategies and consequently does not need to select and configure strategies manually. This eases the mutation testing application and thus reduces the overall cost of the testing activity. For assessing the feasibility of Sentinel, multiple experiments were performed. The experiments were conducted using 10 different systems with 4 versions of each, in a total of 40 systems. The strategies generated by sentinel showed to be significantly more effective than conventional strategies of the literature. Furthermore, they kept their good results in newer versions of the software, which demonstrated their reusability. Keywords: hyper-heuristic, software testing, strategy generation, mutant reduction, grammatical evolution, search based software engineering, multi-objective optimization
koamabayili/VECTRON-author-checklist: VECTRON author checklist
No full textWe have done our best to complete the author checklist relating to the use of animals in the hut study. Note that the objective for the hut study was to evaluate the IRS treatment applications for residual efficacy against Anopheles mosquitoes, including the local An. coluzzii mosquito population. Cows were only used to attract mosquitoes into the huts and no tests were carried out directly on the cows. The author checklist is intended for use with studies where experiments are carried out on animals, which is why we have had such difficulty in completing this for the hut study, as many of the questions do not relate to how the cows were used
- …
