Learning code representations has found many uses in software engineering, such as classification, search, comment generation, and bug prediction, etc. Although of tokens, syntax trees, dependency graphs, paths or the combinations their variants have been proposed, existing learning techniques a major limitation that these models are often trained on datasets labeled for specific downstream tasks, may not be suitable other tasks. Even though some generate from unlabeled code, they far being...

Aspect-oriented programming (AOP) has been attracting much attention in the software engineering community by advocating that programs should be structured according to programmer concerns, such as efficient use of memory. However, like other paradigms their early days, AOP hasn't addressed yet earlier phases development. In particular, it is still an open question how one identifies aspects on development process. This work proposes answer this question. Specifically, we show can discovered...

We introduce a variability-intensive approach to goal decomposition that is tailored support requirements identification for highly customizable software. The based on the semantic characterization of OR-decompositions goals. first show each high-level can be associated with set concerns, in response which, alternative refinements introduced. A text corpus relevant domain discourse used derive such variability concerns are specific problem. In parallel, contextual facts vary while being...

Given the importance of identifier names and value naming conventions to program comprehension, we speculated in previous work whether a connection exists between quality software quality. We found that flawed identifiers Java classes were associated with source code be low by static analysis. This paper extends three directions. First, show association also holds at finer granularity level methods. turn makes it possible to, secondly, apply existing method-level readability metrics, see...

Security bug reports can describe security critical vulnerabilities in software products. Bug tracking systems may contain thousands of reports, where relatively few them are related. Therefore finding unlabelled bugs among be challenging. To help engineers identify these quickly and accurately, text-based prediction models have been proposed. These often mislabel due to a number reasons such as class imbalance, the ratio non-security is very high. More critically, we observed that presence...

The instantaneous coefficient of variation (ICOV) edge detector, based on normalized gradient and Laplacian operators, has been proposed for detection in ultrasound images. In this paper, the localization performance ICOV-squared (ICOVS) detector are examined. First, a simplified version ICOVS magnitude squared, is scrutinized order to reveal statistical speckled imagery. Both probability false alarm evaluated detector. Edge characterized by position peak 3-dB width response. Then,...

Autonomic computing systems reduce software maintenance costs and management complexity by taking on the responsibility for their configuration, optimization, healing, protection. These tasks are accomplished switching at runtime to a different system behaviour - one that is more efficient, secure, stable, etc. while still fulfilling main purpose of system. Thus, identifying objectives system, analyzing alternative ways how these can be met, designing supports all or some behaviours...

Context-aware applications monitor changes in their operating environment and switch behaviour to keep satisfying requirements. Therefore, they must be equipped with the capability detect variations context response such variations. However, specifying monitoring switching can difficult due dependence on varying contextual properties which need made explicit. In this paper, we present a problem- oriented approach represent reason about variability assess its impact requirements; elicit...

Studies have demonstrated the importance of good identifier names to program comprehension. It is unclear, however, whether poor naming has other effects that might impact maintenance effort, e.g. on code quality. We evaluated quality in 8 established open source Java applications libraries, using a set 12 guidelines. found statistically significant associations between flawed identifiers (i.e. violating at least one guideline) and issues reported by FindBugs, static analysis tool.

Autonomic computing systems reduce software maintenance costs and management complexity by taking on the responsibility for their configuration, optimization, healing, protection. These tasks are accomplished switching at runtime to a different system behaviour - one that is more efficient, secure, stable, etc. while still fulfilling main purpose of system. Thus, identifying analyzing alternative ways how objectives can be achieved designing supports all these behaviours promising way...

A reverse engineering process aims at reconstructing high-level abstractions from source code. This paper presents a novel methodology for recovering stakeholder goal models both structured and unstructured legacy The consists of the following major steps: 1) Refactoring code by extracting methods based on comments; 2) Converting refactored into an abstract program through statechart refactoring hammock graph construction; 3) Extracting model program's syntax tree; 4) Identifying...

Goal models are effective in capturing stakeholder needs at the time when features of system-to-be have not yet been conceptualized. Relating goals to solution-oriented gives rise a requirement traceability problem. In this paper, we present new model-driven extension an Early Requirements Engineering tool (OpenOME) that generates initial feature model from goals. Enabled by such generative mapping, configuration constraints among variability can be obtained reasoning about

Monitoring the satisfaction of software requirements and diagnosing what went wrong in case failure is a hard problem that has received little attention Software Requirement Engineering literature. To address this problem, we propose framework adapted from artificial intelligence theories action diagnosis. Specifically, monitors generates log data at level granularity can be tuned adaptively runtime depending on monitored feedback. When errors are found, diagnoses denial identifies...

A self-adaptive system uses runtime models to adapt its architecture the changing requirements and contexts. However, there is no one-to-one mapping between in problem space architectural elements solution space. Instead, one refined requirement may crosscut multiple elements, realization involves complex behavioral or structural interactions manifested as design decisions. In this paper we propose combine two kinds of self-adaptations: requirements-driven self-adaptation, which captures...

Software applications that are very large-scale, can encompass hundreds of complex user interfaces (UIs). Such commonly sold as feature-bloated off-the-shelf products to be used by people with variable needs in the required features and layout preferences. Although many UI adaptation approaches were proposed, several gaps limitations including: extensibility integration legacy systems, still need addressed state-of-the-art adaptive development systems. This paper presents Role-Based...

Algorithm classification is to automatically identify the classes of a program based on algorithm(s) and/or data structure(s) implemented in program. It can be useful for various tasks, such as code reuse, theft detection, and malware detection. Code similarity metrics, basis features extracted from syntax semantics, have been used classify programs. Such features, however, often need manual selection effort are specific individual programming languages, limiting classifiers programs same...