Algorithmic interventions (the intricate, dynamic arrangements of people and code deployed to address everyday life problems) benefit society. However, they are also responsible for many complex social, economic, and political harms. Unfortunately, there is an epistemological gap in computer science. Computer science lacks the language, the knowledge, and the methods for dealing with how to make algorithms’ stakeholders responsible, accountable, and liable for their outcomes’ effects, impacts, and consequences. Our research focuses on seizing pragmatism philosophy and linguistic pragmatics to reform computer science epistemology, expanding SDLC (system development life cycle framework) based on clinical pragmatics. We support reform to SDLC to help diminish algorithmic interventions’ adverse effects by design or, at least, to assign stakeholders’ duties for every algorithmic intervention in society. The algorithmic pragmatism conceptual framework presented could be taught in schools and universities and be used to public regulate and judge algorithmic interventions. Overall, we aim to design means of thinking about computer systems and algorithms to fill the computer science epistemology gap related to the impacts and consequences of algorithmic interventions.

Social machines (SMs) are the term used to define processes in which the people do the creative work and the machine does the administration. The concept was scarcely studied until 2013, when the series of workshops on SMs was created, and the topic began to receive more attention. However, it is not clear how research has evolved since then. This article aims to investigate and summarize how the research field of SM has evolved since 2013, to outline the state of the art and practice, and identify research opportunities within this field. We performed a systematic literature review analyzing the quantity and quality of publications, the main topics addressed, the current classifications of SMs, the context in which the concepts are used, and the main perceived challenges. We identified and analyzed 56 relevant studies addressing 12 topics, representing the current practical landscape of research regarding SM. Our findings suggest that: 1) research interest in SM is increasing, but is still concentrated into two research clusters; 2) topics are grouped under two main headings: a) human behavior and b) software development; 3) there is still a need for a common taxonomy to define and classify SM; 4) the main contexts are crowdsourcing and social networks, and the majority of studies are small-scale studies in an academic setup; and 5) more empirical rigor and evidence is needed regarding their use, benefits and challenges, despite some evidence regarding challenges related to user engagement, trust, scalability, and a better human-machine collaboration. Finally, a vision of the future of SMs, with the integration of web of people, artificial intelligence, and things, is also presented and discussed.

Several approaches have been investigated to develop models able to solve forecasting problems. However, a limitation arises in the particular case of daily-frequency financial forecasting and is called the random walk dilemma (RWD). In this context, the concept of time phase adjustment can be included in forecasting models to overcome such a drawback. But the evolution of trading systems has increased the frequency for performing operations in the stock market for fractions of seconds, which requires the analysis of high-frequency financial time series. Thus, this work proposes a model, called the increasing decreasing linear neuron (IDLN), to forecast high-frequency financial time series from the Brazilian stock market. Furthermore, an evolutionary covariance-based method with automatic time phase adjustment is presented for the design of the proposed model, and the obtained results overcame those obtained by classical forecasting models in the literature.

A wide number of sophisticated models have been proposed in the literature to solve prediction problems. However, a drawback arises in the particular case of financial prediction problems and called the random walk dilemma (RWD). In this context, the concept of time phase adjustment can be used to overcome the problem for daily-frequency financial time series. However, the fast evolution of trading platforms increased the frequency for performing operations in the stock market for fractions of seconds, which makes the analysis of high-frequency financial time series very important in this current scenario. In this way, this paper presents a model, called the increasing decreasing linear neuron (IDLN), to predict high-frequency financial time series from the Brazilian stock market. Besides, a descending gradient-based method with automatic time phase adjustment is presented for the design of the proposed model, and the obtained results overcame those obtained by established prediction models in the literature.

The use of Government Open Data to deliver services and promote transparency is a current goal of many governments. In this paper, we briefly describe two contest winner applications with focus on these goals, and present a discussion about the difficulties of integrating Government Open Data to develop them.

Change Requests (CRs) are key elements to software maintenance and evolution. Finding the appropriate developer to a CR is crucial for obtaining the lowest, economically feasible, fixing time. Nevertheless, assigning CRs is a labor-intensive and time consuming task. In this paper, we present a semi-automated CR assignment approach which combine rule-based and information retrieval techniques. The approach emphasizes the use of contextual information, essential to effective assignments, and puts the development team in control of the assignment rules, toward making its adoption easier. Results of an empirical evaluation showed that the approach is up to 46,5% more accurate than approaches which rely solely on machine learning techniques.

In addition to the benefits brought by the use of Distributed Software Development, new challenges linked to its use also emerged. Due to lack of information companies and organizations around the globe, independently, solve these challenges in many different ways, each with their practices, some more some less efficient, where best practices are hardly widespread among DDS community. In this context, this paper aims to present a web system based on Case Based Reasoning and Natural Language Processing to extract information in text form of problems and solutions adopted by distributed software projects and to recommend similar past experiences in order to support the decisions and resolutions of problems arising from new situations in distributed projects. The feasibility proof of the method was made from experimental tests conducted to identify the success of the recommendations of previous valid cases, with a success rate of 90% for the sample used.

I/O bound systems can significantly reduce theirtotal execution time by performing I/O operationsconcurrently. Despite this enormous potential, most systemsperform I/O operations sequentially. One of the reasons behindthis is that the most widespread mechanisms for concurrentI/O are callback-based, resulting in code hard to write andmaintain. In this context, we propose Afluentes, a Javaframework that allows asynchronous functions to be composedjust as easily as synchronous ones, facilitating the developmentof concurrent I/O intensive systems. We performed anexperimental evaluation whose results showed that Afluentesleads to significant performance gains over sequential I/O.Compared to callback-based approaches, Afluentes leads tosimilar performance gains, but with less programming effort.

ABSTRACT Software maintenance starts as soon as the first artifacts are delivered and is essential for the success of the software. However, keeping maintenance activities and their related artifacts on track comes at a high cost. In this respect, change request (CR) repositories are fundamental in software maintenance. They facilitate the management of CRs and are also the central point to coordinate activities and communication among stakeholders. However, the benefits of CR repositories do not come without issues, and commonly occurring ones should be dealt with, such as the following: duplicate CRs, the large number of CRs to assign, or poorly described CRs. Such issues have led researchers to an increased interest in investigating CR repositories, by considering different aspects of software development and CR management. In this paper, we performed a systematic mapping study to characterize this research field. We analyzed 142 studies, which we classified in two ways. First, we classified the studies into different topics and grouped them into two dimensions:challengesandopportunities. Second, the challenge topics were classified in accordance with an existing taxonomy for information retrieval models. In addition, we investigated tools and services for CR management, to understand whether and how they addressed the topics identified. Copyright © 2013 John Wiley & Sons, Ltd.

The dilation-erosion-linear perceptron (DELP) is a morphological-linear model based on fundamentals of mathematical morphology (MM). Its design is a gradient-based learning process using ideas from the backpropagation (BP) algorithm. However, a drawback arises from the gradient estimation of morphological operators, because they are not differentiable of usual way. In this sense, this paper presents an evolutionary learning process, using the covariance matrix adaptation evolutionary strategy (CMAES), to design the DELP model. Furthermore, we conduct an experimental analysis using a relevant set of binary classification problems, and the obtained results are discussed and compared to results found using the DELP model with its classical learning process.

Background: Along the last decade, there has been a significant increase in the adoption of the approaches based on Distributed Software Development (DSD). This approach has brought several competitive advantages, as well as new challenges such as communication and information sharing. In this context, the ontologies can provide benefits such as the definition, standardization and sharing of knowledge involved in the project, allowing a uniform understanding of information and facilitating the collaboration among distributed software development teams. Aim: Identifying evidence to determine which tools, models, techniques and best practices that use ontologies to support the DSD projects, and which ontologies proposed in this context. Method: This paper presents a systematic mapping study conducted in order to investigate how ontologies are being applied as a support to the DSD. The research protocol was based on Kitchenham's, and Travassos and Biolchini's guidelines. Searches were performed both in manual and automatic way in a set of digital libraries engines and leading conferences in the Software Engineering field. Results: From the initial set of 1588 studies, it was selected a total of 38 primary studies that answer the two research questions. Conclusions: This work presents evidences from each paper collected and an analysis of results reached. The results support the foundation for proposing and developing a feature based on ontologies to support the DSD, besides encouraging further researches that may promote advancements in this area and fostering the adoption of these types of resources by the global software industry.

Context: Change Request (CR) repositories play an important role in the software maintenance and evolution process. Through a CR repository, software changes are reported and assigned to developers. Finding the appropriate developer to a CR is crucial for obtaining the lowest, economically feasible, fixing time. Nevertheless, assigning CRs is a labor-intensive and time consuming task. Although many work have proposed automated approaches for CR assignment, they have been implemented without investigating the fundamental aspects which characterize the task itself. Objective: This paper investigates the effort that is taken to assign CR to appropriate developers and identifies the fundamental aspects that characterize it, such as the strategies to perform the assignments and the complexity involved in them. Such investigation improves the current knowledge on the topic, providing researchers and practitioners with useful information towards developing effective solutions. Method: A survey was performed with software developers to understand CR assignment in the Brazilian Federal Organization for Data Processing. The questionnaire was composed of 38 questions, being them both open-ended and closed-ended. We analyzed the answers of 36 respondents. Results: We find that: there is a significant amount of time being spent on assignments (e.g., assigning 20 CRs can take up to 3.3 hours); there are many strategies used to assign CRs, which are complementary to those used in current automated solutions; and CR assignment is very complexity due to a process that requires cognitive abilities for information seeking, communication, and memorization. Conclusion: CR repositories are fundamental to software maintenance, however assigning CRs to developers is an expensive activity. Although we understand that fully and totally accurate automation of assignments is unlikely, further improvements on this direction are feasible and necessary to reduce costs. This way, this paper brings relevant findings to guide new research on automated CR assignment.

Content sharing services have become immensel popular on the Web. More than 1 billion people use this kind o services to communicate with friends and exchange all sorts o information. In this new context, privacy guarantees are essential guarantees about the potential release of data to unintended recipients and the use of user data by the service provider Although the general public is concerned about privacy question related to unintended audiences, data usage by service provider is still misunderstood. In order to further explore this level o misunderstanding, this work presents the results of a surve: conducted among 900 people with the aim of discovering hov people care about the use of their personal data by servic providers in terms of social media. From the results, we found that: (i) in general people do not read license terms and do no know very much about service policies, and when presented with these policies people do not agree with them; (ii) a good number of people would support alternative models such as paying for privacy or selling their personal data; and (iii) there are some differences between generations in relation to how they care about their data.

Software Product Line (SPL) Engineering focuses on systematic software reuse, which has benefits such as reductions in time-to-market and effort, and improvements in the quality of products. However, establishing a SPL is not a simple matter, and can affect all aspects of the organization, since the approach is complex and involves major investment and considerable risk. These risks can have a negative impact on the expected ROI for an organization, if SPL is not sufficiently managed. This paper presents a mapping study of Risk Management (RM) in SPL Engineering. We analyzed a set of thirty studies in the field. The results points out the need for risk management practices in SPL, due to the little research on RM practices in SPL and the importance of identifying insight on RM in SPL. Most studies simply mention the importance of RM, however the steps for managing risk are not clearly specified. Our findings suggest that greater attention should be given, through the use of industrial case studies and experiments, to improve SPL productivity and ensure its success. This research is a first attempt within the SPL community to identify, classify, and manage risks, and establish mitigation strategies.

Newcomers in a software development project often need assistance to complete their first tasks. Then a mentor, an experienced member of the team, usually teaches the newcomers what they need to complete their tasks. But, to allocate an experienced member of a team to teach a newcomer during a long time is neither always possible nor desirable, because the mentor could be more helpful doing more important tasks. During the development the team interacts with a version control system, bug tracking and mailing lists, and all these tools record data creating the project memory. Recommender systems can use the project memory to help newcomers in some tasks answering their questions, thus in some cases the developers do not need a mentor. In this paper we present Mentor, a recommender system to help newcomers to solve change requests. Mentor uses the Prediction by Partial Matching (PPM) algorithm and some heuristics to analyze the change requests, and the version control data, and recommend potentially relevant source code that will help the developer in the change request solution. We did three experiments to compare the PPM algorithm with the Latent Semantic Indexing (LSI). Using PPM we achieved results for recall rate between 37% and 66.8%, and using LSI the results were between 20.3% and 51.6%.

Background - Risk Management (RM) practices are usually established towards avoiding or minimizing problems, likely to occur during software development. It can be stated as the task of analyzing and managing the impact of every important action to be performed in the project. Aim - In the context of RM practices, we developed a scoping study, aiming at analyzing the current scenario of RM practices in software development. Method - We analyzed 74 studies published by the most important venues published up to the year 2011. Based on the analyzed dataset, we sketched a set of useful practices for applying RM in software projects. Results - The analysis indicate that most of the studies subjectively describe ways to evaluate risks, instead of providing readers with details on how RM is to be performed. Conclusions - Such findings points out to the need of further research in the field of RM, specially due to its importance for software development projects.

Software Product Lines (SPL) adoption can affect several aspects of an organization and it involves significant investment and risk. This way, SPL risk management is a crucial activity of SPL adoption. This study aims to identify SPL risks during the scoping and requirement disciplines to provide information to better understand risk management in SPL. In order to achieve the previous stated goal, a case study research was applied in an industrial project in the medical information management domain. Using the captured risks, a classification scheme was built and risk mitigation strategies were identified. We spent five months, totaling 79 hours, performing risk management (RM) in the scoping discipline and twelve months, totaling 148 hours, performing RM on the requirements discipline. We identified 32 risks during the scoping discipline and 20 risks during the requirements discipline, 14 risks occurred in both disciplines. Some identified risks are not particular to SPL development, however, they have their impact increased due to the SPL characteristic. All the study results and lessons learned are useful for all project managers and researchers who are considering the introduction of SPL risk management in industry or academia.

The global, local and firm economies depend on software in an ever increasing way. Writing effective software on a day to day basis, for a long time (decades, maybe) depends upon agility (as a process and timing) to have, as consequences, reliability, scalability and security, in times of software and systems as services.

The combination of service-orientation and software product line engineering, called Service-Oriented Product Line Engineering (SOPLE) have received attention by researchers and practitioners in the last years, and these areas can address issues of each other. One service-orientation issue is service identification. It consists of determining candidate services to a service-oriented environment based on pre-existing software artifacts, e.g., business process, source code, and so on. In order to provide a systematic identification of services, there are many available service identification methods in the literature, regarding different understanding of services, goals, and techniques. Due to this heterogeneity, this paper presents an in-depth comparison of service identification methods as well as a recommendation of the most suitable ones in the SOPLE context. This work can help the decision making of the most suitable method according to stakeholders' needs.

We define a notion of social machine and envisage an algebra that can describe networks of such. To start with, social machines are defined as tuples of input, output, processes, constraints, state, requests and responses, apart from defining the machines themselves, we define their connectors and conditionals that can be used to describe the interactions between any number of machines in a multitude of ways, as a means to represent real machines interacting in the real web, such as Twitter, Twitter running on top of Amazon AWS, mashups built using Twitter and, obviously, other social machines. This work is not a theoretical paper as yet, but, in more than one sense, we think we have found a way to describe web based information systems and are starting to work on what could be a practical way of dealing with the complexity of this emerging web of social machines that is all around us.

Context: The benefits of taking a product line approach in order to achieve significant reductions in cost and time to market and, at the same time, increasing the quality has encouraged product line adoption. Objective: In this context, this study focuses on some SPL adoption aspects and has the following goals: investigate state-of-the-art SPL adoption, synthesize available evidence, and identify gaps between required strategies, organizational structures, maturity level and existing adoption barriers, available in the literature. Method: A systematic mapping study was undertaken to analyze the important aspects that should be considered when adopting SPL approaches. A set of four questions were defined in which 34 primary studies were evaluated. Results: A total of 34 primary studies were considered. They reported four different strategies (Incremental, Big Bang, Tactical and Pilot project), however there is insufficient information about how such strategies link to factors as organizational structure and process maturity. By investigating all primary studies we found 23 barriers to adoption. Conclusions: Researchers need to consider the relationships between SPL adoption and factors such as company maturity and organization structure in more detail. There is also a need for patterns to assist in SPL adoption and overcoming SPL adoption barriers.

In this work we present an evolutionary learning process using the covariance matrix adaptation evolutionary strategy (CMAES) to design the dilation-erosion perceptron (DEP) for stock market indices forecasting. Also, we have included an automatic phase fix procedure (APFP) into proposed learning process to eliminate time phase distortions observed in some forecasting problems. The main advantage of the DEP model designed by our learning process, apart from its higher forecasting performance, is do not request any methodology to overcome the nondifferentiability of morphological operators needed into classical gradient-based learning process of the DEP model. Besides, we present an experimental analysis using two stock market indices, where five well-known performance metrics and an evaluation function are used to assess forecasting performance.

In this paper we present a gradient-based morphological approach to solve the software development cost estimation (SDCE) problem. The proposed approach consists of a dilation-erosion perceptron (DEP) trained by a gradient steepest descent method using the back propagation (BP) algorithm and a systematic approach to overcome the problem of nondifferentiability of morphological operators. Furthermore, we compare the proposed approach with other neural and statistical models using five complex SDCE problems.

The dilation-erosion perceptron (DEP) is a class of hybrid artificial neurons based on framework of mathematical morphology (MM) with algebraic foundations in the complete lattice theory (CLT). A drawback arises from the gradient estimation of dilation and erosion operators into classical gradient-based learning process of the DEP model, since they are not differentiable of usual way. In this sense, we present a differential evolutionary learning process, called DEP(MDE), using a modified differential evolution (MDE) to design the DEP model for air pressure forecasting. Also, we have included an additional step into learning process, called automatic phase fix procedure (APFP), to eliminate time phase distortions observed in some forecasting problems. Furthermore, an experimental analysis is presented using two complex time series, where five well-known performance metrics and an evaluation function are used to assess forecasting performance.

The Dilation-erosion perceptron (DEP) is considered a good forecasting model, whose foundations are based on mathematical morphology (MM) and complete lattice theory (CLT). However, a drawback arises from the gradient estimation of morphological operators into classical gradient-based learning process, since they are not differentiable of usual way. In this sense, this work presents an evolutionary learning process, called DEP(MGA), using a modified genetic algorithm (MGA) to design the DEP model for weather forecasting. In addition, we have included an automatic phase fix procedure (APFP) into the proposed learning process to eliminate time phase distortions observed in some temporal phenomena. At the end, an experimental analysis is presented using two complex time series, where five well-known performance metrics and an evaluation function are used to assess forecasting performance.

Abstract Background : Software product lines and Agile methods have been an effective solution for dealing with the growing complexity of software and handling competitive needs of software organizations. They also share common goals, such as improving productivity, reducing time‐to‐market, decreasing development costs and increasing customer satisfaction. There has been growing interest in whether the integration of Agile and SPL could provide further benefits and solve many of the outstanding issues surrounding software development. Objective : This study investigates the state‐of‐the‐art in Agile SPL approaches, while identifying gaps in current research and synthesizing available evidence. It also provides a basis for a deeper understanding of the issues involved in the integration of Agile and SPL. Method : A mapping study was undertaken to analyze the relation between Agile and SPL methods. A set of four research questions were defined in which the 32 primary studies were evaluated. Results : This study provides insights into the integration of Agile and SPL approaches, it identifies the current gaps in the research, synthesize the available evidence and propose specific Agile methods and practices for integration in SPL. Conclusions : In general, few studies describe the underlying Agile principles being adopted by proposed Agile SPL solutions. The most common Agile practices proposed by the studies came from the XP and Scrum methods, particularly in the pro‐active SPL strategy. We identify certain Agile methods that are being overlooked by the Agile SPL community, and propose specific SPL practices areas suitable for adoption of Agile practices. Copyright © 2011 John Wiley & Sons, Ltd.

During the information era, the knowledge became a competitive differential. The knowledge management, despite of not been a new subject, has increased its importance. In this scenario, the use of corporative social networks supporting organizations knowledge management initiatives turns out to be a promising approach. This article presents the results of a qualitative analysis about the knowledge management initiative based in social networks made in a Brazilian innovation institute. The analysis points out some problems and suggests some improvements in organizational knowledge management processes definition.

Historically, it is well known that issues related to security of software applications are normally omitted by the development teams owing to a lack of expertise or knowledge in security policies. With the emergence of WEB 2.0 applications and technology many systems became to be ported from the original platform to the WEB, making the security flows became more serious. Entire systems, complex or not, have outstanding access availability and therefore are highly vulnerable to threats. This work aims to discuss how the security requirements should be elaborated in order to making easier the test execution team to elaborate the tests cases and consequently improving the quality of the solution developed.

In terms of storage and processing, is widely known that organizations infrastructure is misused, being idle most of the time. Consequently, there is an emerging need for solutions that take advantages of the idle capacity of this infrastructure. In this context, the P2P architecture proves to be an efficient way to make a good use of this capacity by implementing distributed backup systems. This work presents a proposal for an algorithm that can be used in P2P networks to quantify as well as ensures the availability of restoring files anytime in percentage terms, so that any hardware failure does not affect the backup operation. The algorithm proposed improves reliability on P2P network, making possible to validate using percent values the availability of a peer.

Since the late seventies, efforts to catalog factors that influences productivity, as well as actions to improve it, has been a huge concern for both academy and software development industry. Despite numerous studies, software organizations still do not know which the most significant factors are and what to do with it. Several studies present the factors in a very superficial way, some others address only the related factors or there are those that describe only a single factor. Actions to deal with the factors are spread and frequently were not mapped. Through a literature review, this paper presents a consolidated view of the main factors that have affected productivity over the years, and the strategies to deal with these factors nowadays. This research aims to support software development industry on the selection of their strategies to improve productivity by maximizing the positive factors and minimizing or avoiding the impact of the negative ones.

Institutions, large and small, are increasingly using Internet infrastructure, services and applications to decentralize staff and operations, evolving from a closed to an open house, networked environment, where more and more informal knowledge needs to be shared and managed. At the same time, with a very large percentage of corporate processes being formally supported by information systems, there is an ever increasing gap separating the institutions' explicit and implicit knowledge silos. In this talk, we describe a social network infrastructure currently in use in a number of corporations and the way in which we are integrating the notion of project as an integral part of the system. We also discuss how such combination of formal/informal knowledge management can be an essential feature of collaborative work in corporations and development efforts in general.

The Information Communication and Technology – ICT industry is facing a market of constant changes and challenges. These characteristics demand the ICT professionals to have a wide vision of the problem, rather than just knowledge on the technology. In this context, the objective of this article is to propose an innovative pedagogical methodology based on PBL (Problem Based Learning) to improve the learning effectiveness in software engineering, through the implementation of software factories, where students can work together to solve real problems. This methodology is running in a graduate course in Software Engineering managed by CESAR , a research institute with experience in the development of innovative software. Some important results about this course are presented and discussed in this paper.

Component-Based Software Engineering (CBSE) is being used in a wide variety of application areas, including embedded systems, and the correct operation of the components is often critical for business success and, in some cases, human safety. In this way, assessment and evaluation of software components has become a compulsory and crucial part of any CBSE lifecycle. A risk of selecting a product with unknown quality properties is no longer acceptable and, when happened, may cause catastrophic results. Thus, the software components quality evaluation has become an increasingly essential activity in order to bring reliability in (re)using software components. In this way, we propose a quality verification framework to evaluate the quality of embedded software components in an efficient way. This model is based in component characteristics, sub-characteristics, quality attributes and metrics.

It is historically known that extensive phases of rework affect directly both costs and quality of already developed applications. In regard to requirements, systems security figure among those which are left behind, generating undesired steps of reformulation. This article is a short presentation on design patterns and non functional security requirements; it relates both and shows that an initial use of certain patterns, in early development phases, might reduce the impacts of relegating the systems security requirement implementation to a second plan.

Measuring the productivity in software development projects is a historical complex problem. The main problem resides in quantifying the relevant results produced in a software project. Different organizations and persons have distinguished notion about the concept of what is produced and its associated value. Therefore, the appropriate model for measuring productivity in software organizations needs to be unique to each type of organization. In addition, we encountered, based on a broder investigation about the productivity metrics and measurement studies published in literature, that there seems to be a trend towards the use of what we called Value Size based Metrics for productivity measurement. Based on these findings, we defined a process to support the definition of productivity measurement models in software organizations and which takes into account the difficulties of measuring the productivity and reflects the best practices in measurement programs. The adoption of a productivity measurement model based on value helps the organization to focus its efforts to improve productivity in the direction suitable with the high management vision.

This paper explains how incentive systems are defined in software development organizations and the impacts that they can have on team productivity. It also supplies a set of guidelines which is being applied in an organization for the purposes of helping managers to define and implement a reward program as part of the organizational strategy to increase the productivity of software development project teams. Besides this, it also tackles the negative impacts that these systems can cause on the productivity of teams when they are badly applied, thus generating the effect of the dysfunction of the measurement system.

In order to attend the industry needs, it is necessary to provide more practical issues of real software development in the academic curricula. This paper describes an educational experience in a post-graduation course at Federal University of Pernambuco, Brazil, involving software reuse. The course's goals were to provide the state-of-the-art of software development with reuse, and the state-of-the-practice developing the Software Product Line (SPL) approach based on a real project. Moreover, we describe the SPL process steps and the educational methodology applied, as well as the projectpsilas difficulties and lessons learned, along with possible improvements for the course extracted from studentspsila feedback during the course and the project. We also expect that this work can be used as a reference material for other universities planning on teaching software reuse or SPL courses.

A repository is a necessary prerequisite to support software engineers and other users in the process of developing software for and with reuse. In the literature, there are several works that explore reuse repositories, however their focus is mostly on reusable component search and retrieval issues, while important aspects of reuse repositories have not been properly explored. On the other hand, some questions raised by companies that desire to adopt or build a reuse repository remain unanswered. Motivated by these questions, this paper presents the specification, design and implementation of a reuse repository that was successfully constructed and deployed in real Brazilian software factories.

Component-based software development focuses on reuse of already available software components, reducing development time and cost and improving software quality and productivity. To enable composition, components must conform to a specific component model and target a specific component platform. This paper presents a component model for embedded systems called Flexcm and the execution environment in which the components will be running. The component model supports aspects like reflectivity, exceptions, events and lifecycle management. The execution environment supports update of specific components, registration of new components with new functionalities and adaptability based on its execution context. In order to achieve a concrete result, the presented component model and platform is driven by the development of a middleware submitted as a proposal to the Brazilian Digital TV System.

The precision and reliability of the estimation of the effort of software projects is very important for the competitiveness of software companies. Good estimates play a very important role in the management of software projects. Most methods proposed for effort estimation, including methods based on machine learning, provide only an estimate of the effort for a novel project. In this paper we introduce a method based on machine learning which gives the estimation of the effort together with a confidence interval for it. In our method, we propose to employ robust confidence intervals, which do not depend on the form of probability distribution of the errors in the training set. We report on a number of experiments using two datasets aimed to compare machine learning techniques for software effort estimation and to show that robust confidence intervals can be successfully built.

Software reuse is essential for improving the productivity and quality of software projects. One of the key issues to promote the adoption of software reuse in companies is the development of effective repositories of software components. It is also very important to have good methods for searching and retrieval of the components. Clustering techniques can help by providing a visualization of the repository of software components as well as in helping to refine the searches by grouping together similar components. In this paper we quantitatively compare two clustering techniques, namely, self-organizing maps (SOM) and growing hierarquical SOM (GHSOM) for clustering a repository of classes from a Java API for building mobile systems. The performance measure was the quantization error. The simulations have shown that GHSOM outperforms SOM in these tasks. GHSOM is more suitable for this task because it is a constructive technique, which is an advantage in tackling the growth of the repository of software components.

The precision and reliability of the estimation of the effort of software projects is very important for the competitiveness of software companies. Good estimates play a very important role in the management of software projects. Most methods proposed for effort estimation, including methods based on machine learning, provide only an estimate of the effort for a novel project. In this paper we introduce a method based on machine learning which gives the estimation of the effort together with a confidence interval for it. In our method, we propose to employ robust confidence intervals, which do not depend on the form of probability distribution of the errors in the training set. We report on a number of experiments using two datasets aimed to compare machine learning techniques for software effort estimation and to show that robust confidence intervals for the effort estimation can be successfully built.

Software reuse is essential for improving the productivity and quality of software projects. One of the key issues to promote the adoption of software reuse in companies is the development of effective repositories of software components. It is also very important to have good methods for searching and retrieval of the components. Clustering techniques can help by providing a visualization of the repository of software components as well as in helping to refine the searches by grouping together similar components. In this paper we quantitatively compare two clustering techniques, namely, self-organizing maps (SOM) and growing hierarchical SOM (GHSOM) for clustering a repository of classes from a Java API for building mobile systems. The performance measure was the quantization error. The simulations have shown that GHSOM outperforms SOM in these tasks. GHSOM is more suitable for this task because it is a constructive technique, which is an advantage in tackling the growth of the repository of software components.

Much of the current research in software defect prediction focuses on building classifiers to predict only whether a software module is fault-prone or not. Using these techniques, the effort to test the software is directed at modules that are labelled as fault-prone by the classifier. This paper introduces a novel algorithm based on constructive RBF neural networks aimed at predicting the probability of errors in fault-prone modules; it is called RBF-DDA with Probabilistic Outputs and is an extension of RBF-DDA neural networks. The advantage of our method is that we can inform the test team of the probability of defect in a module, instead of indicating only if the module is fault-prone or not. Experiments carried out with static code measures from well-known software defect datasets from NASA show the effectiveness of the proposed method. We also compared the performance of the proposed method in software defect prediction with kNN and two of its variants, the S-POC-NN and R-POC-NN. The experimental results showed that the proposed method outperforms both S-POC-NN and R-POC-NN and that it is equivalent to kNN in terms of performance with the advantage of producing less complex classifiers.

This paper proposes and investigates the use of bagging predictors to improve performance of regression methods for estimation of the effort to develop software projects. We have applied bagging to M5P/regression trees, M5P/model trees, multi-layer perceptron (MLP), linear regression and support vector regression (SVR). This article reports on the influence of bagging on the performance of each of these regression methods in the estimation of the effort of software projects. Experiments carried out using a dataset of software projects from NASA show that bagging is able to significantly improve performance of regression methods in this task. Moreover, we show that bagging with M5P/model trees considerably outperforms previous results reported in the literature obtained by both linear regression and RBF networks. It is also shown that bagging with M5P/model trees obtains results comparable to those of SVR, with the advantage of producing more interpretable results.

Since the first works involving software reuse, domain engineering is considered a key process to develop reusable and flexible software. However, the results have shown that there is still much to do before the vision of domain engineering is completely achieved. Among the reasons for this problem, we may highlight the lack of a process to support the design of domain-specific software architectures. This paper presents such an approach, based on a well defined set of principles, guidelines and metrics.

Software engineering and reuse-oriented tools have been studied along the last years, aiming to provide help in the software development. With the importance of reuse growing significantly, effective software reuse tools and environments started to be needed. This paper presents and discusses some works that comprise many issues related to source code search tools, covered by university and industry since 90's until today. In the end of the paper, a set of requirements is presented, integrating the features that can be found in most works of the area, serving as a basis for future work toward an effective source code search tool.

A fundamental premise for any type of reuse is the knowledge about the existence of the object to be reused. Such knowledge may already be available, for example, due to the past experience of the subject of the reuse action or may be obtained through knowledge dissemination. Information retrieval is a key mechanism for allowing a uniform dissemination of the knowledge about available reusable objects.The instantiation of this problem to the software reuse field is the subject of this work. The synergy among the software reuse and information retrieval fields is exploited in the formulation, implementation and evaluation of an integrated environment that aims at promoting a greater reuse activity level on the quest for developing software with better quality while consuming fewer resources.

The Dynamic Decay Adjustment (DDA) algorithm is a fast constructive algorithm for training RBF neural networks. This paper proposes a method for improving RBF-DDA generalization performance by adjusting the weights of the connections between hidden and output units. The method proposed here has been evaluated on three optical character recognition datasets from the UCI repository. The results show that the proposed method considerably improves performance of RBF-DDA in these tasks without increasing the size of the networks. The results are compared to MLP, k-NN, AdaBoost and SVM results reported in the literature. It is shown that the proposed method outperforms MLP and AdaBoost and obtains results comparable to k-NN and SVM on these datasets.

Novelty detection in time series is an important problem with application in different domains. such as machine failure detection, fraud detection and auditing. We have previously proposed a method for time series novelty detection based on classification of time series windows by RBF-DDA neural networks. The paper proposes a method to be used in conjunction with this time series novelty detection method whose aim is to improve performance by adequately selecting the window size and the RBF-DDA parameter values. The method was evaluated on six real-world time series and the results obtained show that it greatly improves novelty detection performance.

This paper presents the ActiveSearch Agent, a proactive agent for suggesting similar documents to the one being browsed/edited by the user. It was designed to provide access to different documents bases in the Internet (through meta-searches to available search engines), in Intranets and in local directories, as well as to process different documents' formats (e.g., HTML, DOC, PS). The retrieval process is on-line, and it is based on the document's content and on the user's preferences. The algorithm used to measure similarity between documents is based on Vector Space model. The prototype runs in Windows OS. Preliminary tests demonstrated a very good system's performance with a superiority of 34 percentile points in the precision in relation to the GoogleToolbar.

This article describes the process of developing a system for translating VDM executable specifications into Lazy ML prototypes. The system was specified in VDM and a Lazy ML prototype implementation was derived from its specification. This article concentrates on discussing the lessons learned in each stage of the development process, evaluating the adequacy of the used methodology. © 1997 by John Wiley & Sons, Ltd.