An information retrieval technique, latent semantic indexing, is used to automatically identify traceability links from system documentation program source code. The results of two experiments in existing software systems (i.e., the LEDA library, and Albergate) are presented. These compared with other similar type experimental link identification using different types techniques. method presented proves give good by comparison additionally it a low cost, highly flexible apply regards...

Concept location identifies parts of a software system that implement specific concept originates from the problem or solution domain. is very common engineering activity directly supports maintenance and evolution tasks such as incremental change reverse engineering. This work addresses using an advanced information retrieval method, Latent Semantic Indexing (LSI). LSI used to map concepts expressed in natural language by programmer relevant source code. Results case study on NCSA Mosaic...

An information retrieval technique, latent semantic indexing, is used to automatically identify traceability links from system documentation program source code. The results of two experiments in existing software systems (i.e., the LEDA library, and Albergate) are presented. These compared with other similar type experimental link identification using different types techniques. method presented proves give good by comparison additionally it a low cost, highly flexible apply regards...

Abstract A comprehensive literature survey on approaches for mining software repositories (MSR) in the context of evolution is presented. In particular, this deals with those investigations that examine multiple versions artifacts or other temporal information. taxonomy derived from analysis and presents work via four dimensions: type mined (what), purpose (why), adopted/invented methodology used (how), evaluation method (quality). The demonstrated to be expressive (i.e., capable...

Source code duplication occurs frequently within large software systems. Pieces of source code, functions, and data types are often duplicated in part or whole, for a variety reasons. Programmers may simply be reusing piece via copy paste they "re-inventing the wheel". Previous research on detection clones is mainly focused identifying pieces with similar (or nearly similar) structure. Our approach to examine text (comments identifiers) identify implementations high-level concepts (e.g.,...

Focuses on investigating the combined use of semantic and structural information programs to support comprehension tasks involved in maintenance reengineering software systems. Semantic refers domain-specific issues (both problem development domains) a system. The other dimension, information, such as actual syntactic structure program, along with control data flow that it represents. An advanced retrieval method, latent indexing, is used define similarity measure between components....

The research examines the version histories of nine open source software systems to uncover trends and characteristics how developers commit code control (e.g., subversion). goal is characterize what a typical or normal looks like with respect number files, lines, hunks committed together. results these three are presented commits categorized from extra small large. findings show that approximately 75% quite for examined along all characteristics. Additionally, messages most common words...

An empirical study to determine if identifier-naming conventions (i.e., camelCase and under_score) affect code comprehension is presented. eye tracker used capture quantitative data from human subjects during an experiment. The intent of this replicate a previous published at ICPC 2009 (Binkley et al.) that timed response test method acquire data. use eye-tracking equipment gives additional insight overcomes some limitations traditional gathering techniques. Similarities differences between...

An eye-tracking study is presented that investigates how individuals find defects in source code. This work partially replicates a previous by Uwano et al. [2006]. In the study, eye movements are used to characterize performance of reviewing Their analysis showed subjects who did not spend enough time initially scanning code tend take more finding defects. The here follows similar setup with added measures and analyses on effectiveness efficiency respect gaze. subject pool larger comprised...

SrcML is an XML representation for C/C++/Java source code that forms a platform the efficient exploration, analysis, and manipulation of large software projects. The lightweight format allows round-trip transformation from to srcML back with no loss information or formatting. toolkit consists src2srcml tool robust translation srcml2src querying via XPath, XSLT. In this demonstration guide these features provided along use XPath constructing source-code queries XSLT conducting simple transformations.

The paper presents a new 3D representation for visualizing large software systems. origins of this can be directly traced to the SeeSoft metaphor. This work extends these visualization mechanisms by utilizing third dimension, texture, abstraction mechanism, and supporting manipulation techniques user interfaces. By we better represent higher dimensional data than previous 2D views. An overview our prototype tool its basic functionality is given. Applications method particular engineering...

A number of taxonomies to classify and categorize software visualization systems have been proposed in the past. Most notable are those presented by Price (1993) Roman (1993). While these an accurate representation issues, they somewhat skewed with respect current research areas on visualization. We revisit this important work propose a re-alignments addressing engineering tasks large-scale development maintenance. framework emphasize general understanding analysis during maintenance...

An approach for the semi-automated recovery of traceability links between software documentation and source code is presented. The methodology based on application information retrieval techniques to extract analyze semantic from associated documentation. A semi-automatic process defined proposed methodology. paper advocates use latent indexing (LSI) as supporting technique. Two case studies using existing are presented comparing this with others. show positive results approach, especially...

Eye-tracking equipment is used to assess how well a subject comprehends UML class diagrams. The results of study are presented in which eye movements captured non-obtrusive manner as users performed various comprehension tasks on goal the identify specific characteristics diagrams, such layout, color, and stereotype usage that most effective for supporting given task. Results indicate subjects have variation (i.e., navigate diagram) depending their expertise software-design ability solve...

A lightweight fact extractor is presented that utilizes XML tools, such as XPath and XSLT to extract static information from C++ source code programs. The first converted into an representation, srcML, facilitate the use of a wide variety tools. method deemed because only partial parsing done. Additionally, technique quite robust can be applied incomplete noncompilable code. trade off this approach queries on some low level details cannot directly addressed. This benchmark comparison with...

The srcML toolkit for lightweight transformation and fact-extraction of source code is described. an XML format C/C++/Java code. open that includes the source-to-srcML srcML-to-source translators round-trip reverse engineering freely available. direct use XPath XSLT supported, archive large projects included, a rich set input output formats through command-line interface Applying transformations formulating queries using very convenient. Application use-cases are shown demonstrated to be...

A means to add explicit structure program source code is presented. XML used augment with syntactic information from the parse tree. More importantly, comments and formatting are preserved identified for future use by development environments comprehension tools. The focus construct a document representation in instead of more traditional data code. This type supports programmer centric view rather than compiler view. Our made relevant respect other research on representations trees...

The paper describes the results of applying Latent Semantic Analysis (LSA), an advanced information retrieval method, to program source code and associated documentation. semantic analysis is a corpus based statistical method for inducing representing aspects meanings words passages (of natural language) reflective in their usage. This methodology assessed application domain software components (i.e., its accompanying documentation). Here LSA used as basis cluster components. clustering...

Focuses on investigating the combined use of semantic and structural information programs to support comprehension tasks involved in maintenance reengineering software systems. "Semantic information" refers domain-specific issues (both problem development domains) a system. The other dimension, information, such as actual syntactic structure program, along with control data flow that it represents. An advanced retrieval method, latent indexing, is used define similarity measure between...

An approach to recover/discover traceability links between software artifacts via the examination of a system's version history is presented. A heuristic-based that uses sequential-pattern mining applied commits in repositories for uncovering highly frequent co-changing sets (e.g., source code and documentation). If different types files are committed together with high frequency then there probability they have link them. The evaluated on number versions open system KDE. As validation step,...

The effect of layout in the comprehension design pattern roles UML class diagrams is assessed. This work replicates and extends a previous study using questionnaires but uses an eye tracker to gather additional data. purpose replication more insight into gaze behavior not evident from questionnaire-based methods. Similarities differences between studies are presented. Four patterns examined two schemes context three open source systems. Fifteen participants answered series eight role...