Analyzing the impact of natural language processing over feature location in models

Lapeña, Raúl; Font, Jaime; Pastor, Oscar; Cetina, Carlos

doi:10.1145/3136040.3136052

Cited by 5 publications

(6 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…4) The Domain Term Extraction and Stopword Removal techniques are applied to automatically filter terms in or out. We selected these NLP techniques to homogenize the text of the inputs of the evolutionary algorithm since they obtained the best results in a previous work [94].…”

Section: Natural Language Processingmentioning

confidence: 99%

Utilizing Automatic Query Reformulations as Genetic Operations to Improve Feature Location in Software Models

Pérez

Ziadi²,

Cetina

2022

IIEEE Trans. Software Eng.

Self Cite

View full text Add to dashboard Cite

In the combination of Model-Driven Engineering (MDE) and Search-Based Software Engineering (SBSE), genetic operations are one of the key ingredients. Our work proposes a novel adaptation of automatic query reformulations as genetic operations that leverage the latent semantics of software models (the cornerstone artefact of MDE). We analyze the impact of these reformulation operations in a real-world industrial case study of feature location in models. As baselines, we use: 1) the widespread single-point crossover plus random mutation; and 2) mask crossover plus random mutation, which is the best performer for feature location in models. We also perform a statistical analysis to provide quantitative evidence of the impact of the results and to show that this impact is significant. Our reformulation operations improve the results of the best baseline by 37.73% in recall and 14.08% in precision. These results are relevant for the task of feature location in models (one of the main activities performed during software maintenance and evolution). Furthermore, given that the only requirement to apply our approach is term availability in models, our work opens a new research direction to improve more tasks in MDE such as bug location or requirements traceability.

show abstract

Section: Natural Language Processingmentioning

confidence: 99%

Utilizing Automatic Query Reformulations as Genetic Operations to Improve Feature Location in Software Models

Pérez

Ziadi²,

Cetina

2022

IIEEE Trans. Software Eng.

Self Cite

View full text Add to dashboard Cite

show abstract

“…For this reason, Natural Language Processing (NLP) techniques are used to process both the requirements and the model fragments before applying the encoding. Specifically, the requirements and the model fragments are processed by a combination of NLP techniques defined in [23], which consists of tokenizing, lowercasing, removal of duplicate keywords, syntactical analysis, lemmatization, and stopword removal.…”

Section: Model Fragmentmentioning

confidence: 99%

“…Since both queries and documents are based on natural language, Natural Language Processing (NLP) techniques are used to process them. In fact, NLP has a direct and beneficial impact on the results, so before applying LSI, the queries and the documents are processed by a combination of NLP techniques defined in [23], which consists of tokenizing, lowercasing, removal of duplicate keywords, syntactical analysis, lemmatization, and stopword removal. Then, LSI constructs vector representations of both a user query and a corpus of text documents by encoding them as a term-by-document co-occurrence matrix and analyzes the relationships between those vectors to get a similarity ranking between the query and the documents (see Figure 10).…”

Section: Tlr-ir: Information Retrieval Baselinementioning

confidence: 99%

Traceability Link Recovery between Requirements and Models using an Evolutionary Algorithm Guided by a Learning to Rank Algorithm: Train control and management case

Marcén

Lapeña

Pastor

et al. 2020

Journal of Systems and Software

Self Cite

View full text Add to dashboard Cite

“…For this reason, Natural Language Processing (NLP) techniques are used to process both the model fragments and the queries before applying the encoding. Specifically, the model fragments and the requirements are processed by a combination of NLP techniques defined in [104], which consists of tokenizing, lowercasing, removal of duplicate keywords, syntactical analysis, lemmatization, and stopword removal.…”

Section: Ontological Encodingmentioning

confidence: 99%

“…the TLR-IR approach). Specifically, the queries and the models are processed by a combination of NLP techniques defined in [104], which consists of tokenizing, lowercasing, removal of duplicate keywords, syntactical analysis, lemmatization, and stopword removal.…”

Section: Approaches Under Evaluationmentioning

confidence: 99%

Design of a Machine Learning-based Approach for Fragment Retrieval on Models

Terraza¹,

Cristina²

View full text Add to dashboard Cite

In this work, we propose an evolutionary ontological encoding approach to enable Machine Learning techniques to be used to perform Software Engineering tasks in models. The approach is based on a domain ontology to encode a model and on an Evolutionary Algorithm to optimize the encoding. As a result, the encoded model that is returned by the approach can then be used by Machine Learning techniques to perform Software Engineering tasks such as concept location, traceability link retrieval, reuse, impact analysis, etc. We have evaluated the approach with an industrial case study to recover the traceability link between the requirements and the models through a Machine Learning technique (RankBoost). Our results in terms of recall, precision, and the combination of both (F-measure) show that our approach outperforms the baseline (Latent Semantic Indexing). We also performed a statistical analysis to assess the magnitude of the improvement.

show abstract

Analyzing the impact of natural language processing over feature location in models

Cited by 5 publications

References 26 publications

Utilizing Automatic Query Reformulations as Genetic Operations to Improve Feature Location in Software Models

Utilizing Automatic Query Reformulations as Genetic Operations to Improve Feature Location in Software Models

Traceability Link Recovery between Requirements and Models using an Evolutionary Algorithm Guided by a Learning to Rank Algorithm: Train control and management case

Design of a Machine Learning-based Approach for Fragment Retrieval on Models

Contact Info

Product

Resources

About