Klérisson Vinícius Ribeiro Paixão

On the Use of Execution Trace Alignment for Driving Perfective Changes

Perfective changes in well-established software systems are easier to perform when the development team has a solid understanding of the internals. However, it is reasonable to assume that the use of an open source system to incorporate new features and obtain a new software product is an appealing approach instead of coding a new product from scratch. Considering this scenario, and considering that it is not uncommon that systems are poorly documented, there is no widely accepted approach to guide the perfective maintenance for developers with low understanding of the system. This work proposes a new method based on the analysis of execution traces for locating evolution points in the source code where changes should be performed. The proposed method was evaluated with three open source systems and the conclusion suggests a significant impact on effort reduction.

Visual Perception Similarities to Improve the Quality of User Cold Start Recommendations

Recommender systems are well-know for taking advantage of available personal data to provide us information that best fit our interests. However, even after the explosion of social media on the web, hence personal information, we are still facing new users without any information. This problem is known as user cold start and is one of the most challenging problems in this field. We propose a novel approach, VP-Similarity, based on human visual attention for addressing this problem. Our algorithm computes visual perceptions similarities among users to build a visual perception network. Then, this networked information is provided to recommender system to generate recommendations. Experimental results validated that VP-Similarity achieves high-quality ranking results for user cold start recommendation.

A Multi-Armed Bandit Model Selection for Cold-Start User Recommendation

How can we effectively recommend items to a user about whom we have no information? This is the problem we focus on in this paper, known as the cold-start problem. In most existing works, the cold-start problem is handled through the use of many kinds of information available about the user. However, what happens if we do not have any information? Recommender systems usually keep a substantial amount of prediction models that are available for analysis. Moreover, recommendations to new users yield uncertain returns. Assuming that a number of alternative prediction models is available to select items to recommend to a cold user, this paper introduces a multi-armed bandit based model selection, named PdMS. In comparison with three baselines, PdMS improves the performance as measured by the nDCG. These improvements are demonstrated on real, public datasets.

Software Evolution Aided by Execution Trace Alignment

Several attempts to facilitate understanding the behavior of software systems have been proposed. Nonetheless, there is no widely accepted approach to facilitate understanding software systems with poor documentation with the goal that new developers could contribute with the evolution of these systems. The effort to understand complex systems may be a prohibitive factor in program comprehension tasks for many developers. This work proposes a new method based on the analysis of execution traces for locating points in the source code where changes that introduce new functionality should be performed. The proposed method was evaluated with a real world system, where new functionality were inserted with low effort.

Processing heterogeneous collections in XML information retrieval

Our model is based on the observation that the tags used in XML documents are semantically related to the content that they delimit. To evaluate the performance of our approach, we participated in the INEX 2004 heterogeneous track, along with 34 other institutions, from which only 5 groups, including us, submitted runs. In this paper we describe how the approach we used in INEX 2004 and 2005 processes heterogeneous collections without any mapping of DTDs.

On the interplay between non-functional requirements and builds on continuous integration

Continuous Integration (CI) implies that a whole developer team works together on the mainline of a software project. CI systems automate the builds of a software. Sometimes a developer checks in code, which breaks the build. A broken build might not be a problem by itself, but it has the potential to disrupt co-workers, hence it affects the performance of the team. In this study, we investigate the interplay between non-functional requirements (NFRs) and builds statuses from 1,283 software projects. We found significant differences among NFRs related-builds statuses. Thus, tools can be proposed to improve CI with focus on new ways to prevent failures into CI, specially for efficiency and usability related builds. Also, the time required to put a broken build back on track indicates a bimodal distribution along all NFRs, with higher peaks within a day and lower peaks in six weeks. Our results suggest that more planned schedule for maintainability for Ruby, and for functionality and reliability for Java would decrease delays related to broken builds.

Recommending source code locations for system specific transformations

From time to time, developers perform sequences of code transformations in a systematic and repetitive way. This may happen, for example, when introducing a design pattern in a legacy system: similar classes have to be introduced, containing similar methods that are called in a similar way. Automation of these sequences of transformations has been proposed in the literature to avoid errors due to their repetitive nature. However, developers still need support to identify all the relevant code locations that are candidate for transformation. Past research showed that these kinds of transformation can lag for years with forgotten instances popping out from time to time as other evolutions bring them into light. In this paper, we evaluate three distinct code search approaches (“structural”, based on Information Retrieval, and AST based algorithm) to find code locations that would require similar transformations. We validate the resulting candidate locations from these approaches on real cases identified previously in literature. The results show that looking for code with similar roles, e.g., classes in the same hierarchy, provides interesting results with an average recall of 87% and in some cases the precision up to 70%.

Preference-Like Score to Cope with Cold-Start User in Recommender Systems

In recent years, there has been an explosion of social recommender systems (SRS) research. However, the dominant trend of these studies has been towards designing new prediction models. The typical approach is to use social information to build those models for each new user. Due to the inherent complexity of this prediction process, for full cold-start user in particular, the performance of most SRS fall a great deal. We, rather, propose that new users are best served by models already built in system. Selecting a prediction model from a set of strong linked users might offer better results than building a personalized model for full cold-start users. We contribute to this line of work comparing several matrix factorization based SRS under full cold-start user scenario, and proposing a general model selection approach, called ToSocialRec, that leverages existing recommendation models to offer items for new users. Our framework is not only able to handle several social network connection weight metrics, but any metric that can be correlated with preference similarity among users, named here as Preference-like score. We perform experiments on real life datasets that show this technique is as efficient or more than current state-of-the-art techniques for cold-start user. Our framework has also been designed to be easily deployed and leveraged by developers to help create a new wave of SRS.

VP-Rec: A Hybrid Image Recommender Using Visual Perception Network

A requirement for a great user experience is to meet the exact needs for the usage of a recommender system. Such systems need users historical preferences to reasonably perform, which might not be the case for a cold-start user. This paper presents VP-Rec, a hybrid image recommender system that addresses the new user cold-start problem. VP-Rec combines user visual perception and pairwise preferences as source of information to perform recommendations. First, we infer pairwise preferences from users ratings. Next, we build visual perception networks linking users according to their visual attention similarities. From these two inferred structures, we build consensual prediction models, so that when a new user enters the system, we capture his visual attention and choose the best model that fits him. The system has been tested on two image datasets, getting important improvements in terms of ranking quality (nDCG) when applied to new user cold-start scenario against state-of-art recommender systems.

Redocumenting APIs with crowd knowledge: a coverage analysis based on question types

BackgroundSoftware libraries and frameworks play an important role in software system development. The appropriate usage of their functionalities/components through their APIs, however, is a challenge for developers. Usually, API documentation, when it exists, is insufficient to assist them in their programming tasks. There are few API documentation writers for the many potential readers, resulting in the lack of explanations and examples concerning different scenarios and perspectives. The interaction of developers on the Web, on the other hand, generates content concerning APIs from different perspectives, which can be used to document APIs, also known as crowd documentation.MethodsIn this paper, we present a study regarding the knowledge generated by the crowd on the Stack Overflow question-and-answer website. Our main goal is to understand how the crowd can contribute for API documentation on two programming tasks: how to implement a scenario using an API (how-to-do-it), and how to fix domain-independent bugs in an existing code where there was a misunderstanding regarding the usage of an API (debug-corrective). We classified questions available on Stack Overflow by the main concerns of askers, and we used those classified as how-to-do-it and debug-corrective to analyze the coverage of API elements on the discussions related to such questions. Our cases included the well-known and popular Swing and Android APIs.ResultsOur main findings showed that the crowd provides more content for debug-corrective tasks than for how-to-do-it tasks, regardless of the API. Android API elements are more discussed by the crowd compared to Swing. Moreover, we observed that some API elements are frequently mentioned together in discussions, and that there is a strong association between API coverage on Stack Overflow and its usage in real software systems.ConclusionsCrowd documentation may not be a complete substitute for official documentation because of its partial coverage, especially for how-to-do-it tasks. However, it can still significantly enhance the existent documentation, especially for the most commonly used API elements, providing code samples and explanations on a large variety of usage nuances. Finally, taking advantage of the high coverage for debug-corrective tasks, a new kind of debugging assistant may be conceived.