Predicting Accurate and Actionable Static Analysis Warnings: An Experimental Approach

Static analysis tools report software defects that may or may not be detected by other verification methods. Two challenges complicating the adoption of these tools are spurious false positive warnings and legitimate warnings that are not acted on. This paper reports automated support to help address these challenges using logistic regression models that predict the foregoing types of warnings from signals in the warnings and implicated code. Because examining many potential signaling factors in large software development settings can be expensive, we use a screening methodology to quickly discard factors with low predictive power and cost-effectively build predictive models. Our empirical evaluation indicates that these models can achieve high accuracy in predicting accurate and actionable static analysis warnings, and suggests that the models are competitive with alternative models built without screening.

Essays on Bayesian Inference for Social Networks

This thesis presents Bayesian solutions to inference problems for three types of social network data structures: a single observation of a social network, repeated observations on the same social network, and repeated observations on a social network developing through time. A social network is conceived as being a structure consisting of actors and their social interaction with each other. A common conceptualisation of social networks is to let the actors be represented by nodes in a graph with edges between pairs of nodes that are relationally tied to each other according to some definition. Statistical analysis of social networks is to a large extent concerned with modelling of these relational ties, which lends itself to empirical evaluation. The first paper deals with a family of statistical models for social networks called exponential random graphs that takes various structural features of the network into account. In general, the likelihood functions of exponential random graphs are only known up to a constant of proportionality. A procedure for performing Bayesian inference using Markov chain Monte Carlo (MCMC) methods is presented. The algorithm consists of two basic steps, one in which an ordinary Metropolis-Hastings up-dating step is used, and another in which an importance sampling scheme is used to calculate the acceptance probability of the Metropolis-Hastings step. In paper number two a method for modelling reports given by actors (or other informants) on their social interaction with others is investigated in a Bayesian framework. The model contains two basic ingredients: the unknown network structure and functions that link this unknown network structure to the reports given by the actors. These functions take the form of probit link functions. An intrinsic problem is that the model is not identified, meaning that there are combinations of values on the unknown structure and the parameters in the probit link functions that are observationally equivalent. Instead of using restrictions for achieving identification, it is proposed that the different observationally equivalent combinations of parameters and unknown structure be investigated a posteriori. Estimation of parameters is carried out using Gibbs sampling with a switching devise that enables transitions between posterior modal regions. The main goal of the procedures is to provide tools for comparisons of different model specifications. Papers 3 and 4, propose Bayesian methods for longitudinal social networks. The premise of the models investigated is that overall change in social networks occurs as a consequence of sequences of incremental changes. Models for the evolution of social networks using continuos-time Markov chains are meant to capture these dynamics. Paper 3 presents an MCMC algorithm for exploring the posteriors of parameters for such Markov chains. More specifically, the unobserved evolution of the network in-between observations is explicitly modelled thereby avoiding the need to deal with explicit formulas for the transition probabilities. This enables likelihood based parameter inference in a wider class of network evolution models than has been available before. Paper 4 builds on the proposed inference procedure of Paper 3 and demonstrates how to perform model selection for a class of network evolution models.

La valutazione degli effetti delle politiche di coesione dell'Unione europea sulla crescita regionale

This PhD thesis aims at providing an evaluation of EU Cohesion policy impact on regional growth. It employs methodologies and data sources never before applied for this purpose. Main contributions to the literature concerning EU regional policy effectiveness have been extensively analysed. Moreover, having carried out an overview of the current literature on Cohesion Policy, we deduce that this work introduces innovative features in the field. The work enriches the current literature with regards to two aspects. The first aspect concerns the use of the instrument of Regression Discontinuity Design in order to examine the presence of a different outcome in terms of growth between Objectives 1 regions and non-Objective 1 regions at the cut-off point (75 percent of EU-15 GDP per capita in PPS) during the two programming periods, 1994-1999 and 2000-2006. The results confirm a significant difference higher than 0.5 percent per year between the two groups. The other empirical evaluation regards the study of a cross-section regression model based on the convergence theory that analyses the dependence relation between regional per capita growth and EU Cohesion policy expenditure in several fields of interventions. We have built a very fine dataset of spending variables (certified expenditure), using sources of data directly provided from the Regional Policy Directorate of the European Commission.

Analisi e ottimizzazione del servizio di raccolta e trasporto dei rifiuti anche in relazione alle tematiche ambientali: un'applicazione alla rete urbana di Bologna

Questo studio, che è stato realizzato in collaborazione con Hera, è un'analisi della gestione dei rifiuti a Bologna. La ricerca è stata effettuata su diversi livelli: un livello strategico il cui scopo è quello di identificare nuovi metodi per la raccolta dei rifiuti in funzione delle caratteristiche del territorio della città, un livello analitico che riguarda il miglioramento delle applicazioni informatiche di supporto, e livello ambientale che riguarda il calcolo delle emissioni in atmosfera di veicoli adibiti alla raccolta e al trasporto dei rifiuti. innanzitutto è stato necessario studiare Bologna e lo stato attuale dei servizi di raccolta dei rifiuti. È incrociando questi componenti che in questi ultimi tre anni sono state effettuate modifiche nel settore della gestione dei rifiuti. I capitoli seguenti sono inerenti le applicazioni informatiche a sostegno di tali attività: Siget e Optit. Siget è il programma di gestione del servizio, che attualmente viene utilizzato per tutte le attività connesse alla raccolta di rifiuti. È un programma costituito da moduli diversi, ma di sola la gestione dati. la sperimentazione con Optit ha aggiunto alla gestione dei dati la possibilità di avere tali dati in cartografia e di associare un algoritmo di routing. I dati archiviati in Siget hanno rappresentato il punto di partenza, l'input, e il raggiungimento di tutti punti raccolta l'obiettivo finale. L'ultimo capitolo è relativo allo studio dell'impatto ambientale di questi percorsi di raccolta dei rifiuti. Tale analisi, basata sulla valutazione empirica e sull'implementazione in Excel delle formule del Corinair mostra la fotografia del servizio nel 2010. Su questo aspetto Optit ha fornito il suo valore aggiunto, implementando nell'algoritmo anche le formule per il calcolo delle emissioni.

Portfolio Approaches in Constraint Programming

Recent research has shown that the performance of a single, arbitrarily efficient algorithm can be significantly outperformed by using a portfolio of —possibly on-average slower— algorithms. Within the Constraint Programming (CP) context, a portfolio solver can be seen as a particular constraint solver that exploits the synergy between the constituent solvers of its portfolio for predicting which is (or which are) the best solver(s) to run for solving a new, unseen instance. In this thesis we examine the benefits of portfolio solvers in CP. Despite portfolio approaches have been extensively studied for Boolean Satisfiability (SAT) problems, in the more general CP field these techniques have been only marginally studied and used. We conducted this work through the investigation, the analysis and the construction of several portfolio approaches for solving both satisfaction and optimization problems. We focused in particular on sequential approaches, i.e., single-threaded portfolio solvers always running on the same core. We started from a first empirical evaluation on portfolio approaches for solving Constraint Satisfaction Problems (CSPs), and then we improved on it by introducing new data, solvers, features, algorithms, and tools. Afterwards, we addressed the more general Constraint Optimization Problems (COPs) by implementing and testing a number of models for dealing with COP portfolio solvers. Finally, we have come full circle by developing sunny-cp: a sequential CP portfolio solver that turned out to be competitive also in the MiniZinc Challenge, the reference competition for CP solvers.

Increasing intercultural competence and tolerance in multicultural schools: A Training Program and its effectiveness

This study reports the implementation of a Training of Intercultural Competence and Tolerance (TICT) for upper-secondary school students and the empirical evaluation of its effectiveness. The TICT program was developed to counteract increasing interethnic conflicts in the North Caucasus Federal District of Russia. It is based on the theoretical and empirical framework of social psychology and cross-cultural psychology. The training effectiveness was assessed by conducting pre- and post-surveys among the training participants. The results indicate that TICT contributes to the development of a positive ethnic identity and the formation of a civic identity among the participating youth. It also increases their optimism regarding the future of interethnic relations in Russia and the subjective level of intercultural competence of majority group youth towards minority cultures. Thus, the evaluation of the training effectiveness of the TICT has shown that the aims of the training have been achieved to a large extent and that the Training of Intercultural Competence and Tolerance can be effectively used to prevent interethnic conflicts and promote interethnic relations in multicultural schools. Suggestions for the practical implementation of the TICT as well as for future research on the training's effectiveness are discussed.

Materials properties from electron density distributions: Molecular magnetism and linear optical properties

The general goal of this thesis is correlating observable properties of organic and metal-organic materials with their ground-state electron density distribution. In a long-term view, we expect to develop empirical or semi-empirical approaches to predict materials properties from the electron density of their building blocks, thus allowing to rationally engineering molecular materials from their constituent subunits, such as their functional groups. In particular, we have focused on linear optical properties of naturally occurring amino acids and their organic and metal-organic derivatives, and on magnetic properties of metal-organic frameworks. For analysing the optical properties and the magnetic behaviour of the molecular or sub-molecular building blocks in materials, we mostly used the more traditional QTAIM partitioning scheme of the molecular or crystalline electron densities, however, we have also investigated a new approach, namely, X-ray Constrained Extremely Localized Molecular Orbitals (XC-ELMO), that can be used in future to extracted the electron densities of crystal subunits. With the purpose of rationally engineering linear optical materials, we have calculated atomic and functional group polarizabilities of amino acid molecules, their hydrogen-bonded aggregates and their metal-organic frameworks. This has enabled the identification of the most efficient functional groups, able to build-up larger electric susceptibilities in crystals, as well as the quantification of the role played by intermolecular interactions and coordinative bonds on modifying the polarizability of the isolated building blocks. Furthermore, we analysed the dependence of the polarizabilities on the one-electron basis set and the many-electron Hamiltonian. This is useful for selecting the most efficient level of theory to estimate susceptibilities of molecular-based materials. With the purpose of rationally design molecular magnetic materials, we have investigated the electron density distributions and the magnetism of two copper(II) pyrazine nitrate metal-organic polymers. High-resolution X-ray diffraction and DFT calculations were used to characterize the magnetic exchange pathways and to establish relationships between the electron densities and the exchange-coupling constants. Moreover, molecular orbital and spin-density analyses were employed to understand the role of different magnetic exchange mechanisms in determining the bulk magnetic behaviour of these materials. As anticipated, we have finally investigated a modified version of the X-ray constrained wavefunction technique, XC-ELMOs, that is not only a useful tool for determination and analysis of experimental electron densities, but also enables one to derive transferable molecular orbitals strictly localized on atoms, bonds or functional groups. In future, we expect to use XC-ELMOs to predict materials properties of large systems, currently challenging to calculate from first-principles, such as macromolecules or polymers. Here, we point out advantages, needs and pitfalls of the technique. This work fulfils, at least partially, the prerequisites to understand materials properties of organic and metal-organic materials from the perspective of the electron density distribution of their building blocks. Empirical or semi-empirical evaluation of optical or magnetic properties from a preconceived assembling of building blocks could be extremely important for rationally design new materials, a field where accurate but expensive first-principles calculations are generally not used. This research could impact the community in the fields of crystal engineering, supramolecular chemistry and, of course, electron density analysis.

Internet-based self-help for schizophrenia – risks and chances

Purpose: Schizophrenia is a severe mental disorder which is accompanied by an enormous individual and societal burden. Despite established efficacy of cognitive behavioral therapy (CBT) for schizophrenia, its dissemination into routine mental health care remains poor. Internet-based cognitive behavioral therapy in a self-help format helps to narrow the treatment gap in many mental disorders. Are Internet-based self-help programs, which are based on the principles of CBT, also feasible and viable for patients with schizophrenia? Methods: Mental health professionals (target N=50) as well as individuals with schizophrenia spectrum disorders (target N=50) reported their opinion regarding potential chances and risks of Internet-based self-help for schizophrenia in an online survey. Results: The preliminary data analysis of n=30 health professionals revealed a general acceptance of Internet-based programs for schizophrenia (53% acceptable, 47% acceptable after empirical evaluation) and specific contraindications (e.g., severe psychotic symptoms; 73%). People with schizophrenia highlighted the attractiveness of self-help interventions due to a wish for empowerment and for opportunities to strengthen self-efficacy. Conclusions: Risks, limitations and chances of Internet-based programs for patients with schizophrenia will be discussed.

Multi-dimensional classification with super-classes

The multi-dimensional classification problem is a generalisation of the recently-popularised task of multi-label classification, where each data instance is associated with multiple class variables. There has been relatively little research carried out specific to multi-dimensional classification and, although one of the core goals is similar (modelling dependencies among classes), there are important differences; namely a higher number of possible classifications. In this paper we present method for multi-dimensional classification, drawing from the most relevant multi-label research, and combining it with important novel developments. Using a fast method to model the conditional dependence between class variables, we form super-class partitions and use them to build multi-dimensional learners, learning each super-class as an ordinary class, and thus explicitly modelling class dependencies. Additionally, we present a mechanism to deal with the many class values inherent to super-classes, and thus make learning efficient. To investigate the effectiveness of this approach we carry out an empirical evaluation on a range of multi-dimensional datasets, under different evaluation metrics, and in comparison with high-performing existing multi-dimensional approaches from the literature. Analysis of results shows that our approach offers important performance gains over competing methods, while also exhibiting tractable running time.

RDSZ: an approach for lossless RDF stream compression

In many applications (like social or sensor networks) the in- formation generated can be represented as a continuous stream of RDF items, where each item describes an application event (social network post, sensor measurement, etc). In this paper we focus on compressing RDF streams. In particular, we propose an approach for lossless RDF stream compression, named RDSZ (RDF Differential Stream compressor based on Zlib). This approach takes advantage of the structural similarities among items in a stream by combining a differential item encoding mechanism with the general purpose stream compressor Zlib. Empirical evaluation using several RDF stream datasets shows that this combi- nation produces gains in compression ratios with respect to using Zlib alone.

Query Rewriting Optimisation Techniques for Ontology-Based Data Access

Ontology-Based Data Access (OBDA) permite el acceso a diferentes tipos de fuentes de datos (tradicionalmente bases de datos) usando un modelo más abstracto proporcionado por una ontología. La reescritura de consultas (query rewriting) usa una ontología para reescribir una consulta en una consulta reescrita que puede ser evaluada en la fuente de datos. Las consultas reescritas recuperan las respuestas que están implicadas por la combinación de los datos explicitamente almacenados en la fuente de datos, la consulta original y la ontología. Al trabajar sólo sobre las queries, la reescritura de consultas permite OBDA sobre cualquier fuente de datos que puede ser consultada, independientemente de las posibilidades para modificarla. Sin embargo, producir y evaluar las consultas reescritas son procesos costosos que suelen volverse más complejos conforme la expresividad y tamaño de la ontología y las consultas aumentan. En esta tesis exploramos distintas optimizaciones que peuden ser realizadas tanto en el proceso de reescritura como en las consultas reescritas para mejorar la aplicabilidad de OBDA en contextos realistas. Nuestra contribución técnica principal es un sistema de reescritura de consultas que implementa las optimizaciones presentadas en esta tesis. Estas optimizaciones son las contribuciones principales de la tesis y se pueden agrupar en tres grupos diferentes: -optimizaciones que se pueden aplicar al considerar los predicados en la ontología que no están realmente mapeados con las fuentes de datos. -optimizaciones en ingeniería que se pueden aplicar al manejar el proceso de reescritura de consultas en una forma que permite reducir la carga computacional del proceso de generación de consultas reescritas. -optimizaciones que se pueden aplicar al considerar metainformación adicional acerca de las características de la ABox. En esta tesis proporcionamos demostraciones formales acerca de la corrección y completitud de las optimizaciones propuestas, y una evaluación empírica acerca del impacto de estas optimizaciones. Como contribución adicional, parte de este enfoque empírico, proponemos un banco de pruebas (benchmark) para la evaluación de los sistemas de reescritura de consultas. Adicionalmente, proporcionamos algunas directrices para la creación y expansión de esta clase de bancos de pruebas. ABSTRACT Ontology-Based Data Access (OBDA) allows accessing different kinds of data sources (traditionally databases) using a more abstract model provided by an ontology. Query rewriting uses such ontology to rewrite a query into a rewritten query that can be evaluated on the data source. The rewritten queries retrieve the answers that are entailed by the combination of the data explicitly stored in the data source, the original query and the ontology. However, producing and evaluating the rewritten queries are both costly processes that become generally more complex as the expressiveness and size of the ontology and queries increase. In this thesis we explore several optimisations that can be performed both in the rewriting process and in the rewritten queries to improve the applicability of OBDA in real contexts. Our main technical contribution is a query rewriting system that implements the optimisations presented in this thesis. These optimisations are the core contributions of the thesis and can be grouped into three different groups: -optimisations that can be applied when considering the predicates in the ontology that are actually mapped to the data sources. -engineering optimisations that can be applied by handling the process of query rewriting in a way that permits to reduce the computational load of the query generation process. -optimisations that can be applied when considering additional metainformation about the characteristics of the ABox. In this thesis we provide formal proofs for the correctness of the proposed optimisations, and an empirical evaluation about the impact of the optimisations. As an additional contribution, part of this empirical approach, we propose a benchmark for the evaluation of query rewriting systems. We also provide some guidelines for the creation and expansion of this kind of benchmarks.

A general-purpose tunable landscape generator

The research literature on metalieuristic and evolutionary computation has proposed a large number of algorithms for the solution of challenging real-world optimization problems. It is often not possible to study theoretically the performance of these algorithms unless significant assumptions are made on either the algorithm itself or the problems to which it is applied, or both. As a consequence, metalieuristics are typically evaluated empirically using a set of test problems. Unfortunately, relatively little attention has been given to the development of methodologies and tools for the large-scale empirical evaluation and/or comparison of metaheuristics. In this paper, we propose a landscape (test-problem) generator that can be used to generate optimization problem instances for continuous, bound-constrained optimization problems. The landscape generator is parameterized by a small number of parameters, and the values of these parameters have a direct and intuitive interpretation in terms of the geometric features of the landscapes that they produce. An experimental space is defined over algorithms and problems, via a tuple of parameters for any specified algorithm and problem class (here determined by the landscape generator). An experiment is then clearly specified as a point in this space, in a way that is analogous to other areas of experimental algorithmics, and more generally in experimental design. Experimental results are presented, demonstrating the use of the landscape generator. In particular, we analyze some simple, continuous estimation of distribution algorithms, and gain new insights into the behavior of these algorithms using the landscape generator.

Adolescents' reactions to universal and indicated prevention programs for depression: Perceived stigma and consumer satisfaction

There is a common view that one of the major considerations in selecting between universal and indicated interventions is the marked stigma produced by the latter. However, to date there has been no empirical examination of this assumption. The current study examined reported stigma and program satisfaction following two school-based interventions aimed at preventing depression in 532 middle adolescents. The interventions were conducted either across entire classes by classroom teachers (universal delivery) or in small high risk groups by mental health professionals (indicated delivery). The indicated delivery was associated with significantly greater levels of perceived stigma, but effect sizes were small and neither program was associated with marked stigma in absolute terms. Perceived stigma was more strongly associated with aspects of the individual including being male and showing greater externalizing symptomatology. In contrast, the indicated program was evaluated more positively by both participants and program leaders and effect sizes for these measures of satisfaction were moderate to large. The results point to the need for further empirical evaluation of both perceived stigma and program satisfaction in providing balanced considerations of the value of indicated and universal programs.

ROTATIVIDADE DOCENTE EM PROGRAMAS DE PÓS-GRADUAÇÃO EM ADMINISTRAÇÃO: UM ESTUDO DE CASOS MÚLTIPLOS

Esta pesquisa analisa a rotatividade docente como uma variável dependente do contexto em que os PPGAs (Programas de Pós-Graduação em Administração) analisados estão imersos. Portanto, foi adotada a Metodologia de Estudo de Casos Múltiplos com viés em redes sociais. Considera-se que as deliberações regulatórias da CAPES (Coordenação de Aperfeiçoamento de Pessoal de Nível Superior) e símbolos tais como estatutos, regimentos e cânones dos casos analisados neste estudo, influenciam as decisões de demitir. Foram encontrados indícios de que se trata da crença em uma rotatividade funcional como prática emergente no campo das IESs confessionais (UMESP, PUC/SP e UPM) como um recurso para adaptar-se às mudanças propostas pela CAPES. Para tal, com objetivo de produzir melhoras na produtividade científica e enquadrar-se aos critérios de avaliação da CAPES, os coordenadores de PPGA-Ego atribuem à rotatividade docente em outro PPGA-Alter como principal fator de sucesso do PPGA-Alter. As evidências encontradas se fundamentam na inter-subjetividade entre coordenadores de distintos programas dos casos analisados, portanto a rotatividade funcional pode ser dependente do contexto e não se trata de em fenômeno aleatório ou mesmo atomístico. O presente trabalho também contribui para a sugestão de futuros trabalhos, como por exemplo, a rotatividade disfuncional além de outros descritos no final. Em todos os três casos PPGA UMESP, PPGA UPM e PPGA PUC-SP ocorreu mobilidade docente para instituições estatais e particulares, que segundo os coordenadores entrevistados representava um movimento desfavorável sob algum aspecto para os docentes que se demitiram. Neste sentido passa a ser necessário um trabalho específico, eventualmente uma avaliação empírica com base nos constructos de intenção em demitir-se fundamentada nos estudos de Comportamento Organizacional ou Psicologia I/O (Industrial e Organizacional) como Congruência Pessoa-Organização (ARGYRIS, 1973, KRISTOFF, 1996), Modelos de RH (ARTHUR, 1982; MOBLEY, 1982; BAUM, 1993), Modelo Steers e Mowday e outros. Porém contextualizado e estruturado.

AstonCAT-plus:an efficient specialist for the TAC market design tournament

This paper describes the strategies used by AstonCAT-Plus, the post-tournament version of the specialist designed for the TAC Market Design Tournament 2010. It details how AstonCATPlus accepts shouts, clears market, sets transaction prices and charges fees. Through empirical evaluation, we show that AstonCAT-Plus not only outperforms AstonCAT (tournament version) significantly but also achieves the second best overall score against some top entrants of the competition. In particular, it achieves the highest allocative efficiency, transaction success rate and average trader profit among all the specialists in our controlled experiments.