Software Component Retrieval Using Rough Sets

Moiz, Salman Abdul

doi:10.1007/978-981-10-8848-3_34

Cited by 1 publication

(1 citation statement)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…After obtaining the assessment and decision table of software project risk factors, the rough set theory was adopted for risk attributes reduction, and the virtue of the rough set software ROSETTA, developed collaboratively by the Norwegian University of Science and Technology and Warsaw University was used [22]. The frequently-used reduction methods, the genetic algorithm and Johnson's algorithm, were adopted for learning sample data reduction.…”

Section: Data Preprocessing and Attribute Deductionmentioning

confidence: 99%

Support or Risk? Software Project Risk Assessment Model Based on Rough Set Theory and Backpropagation Neural Network

Jiang

Hsu

et al. 2019

Sustainability

View full text Add to dashboard Cite

Software supports continuous economic growth but has risks of uncertainty. In order to improve the risk-assessing accuracy of software project development, this paper proposes an assessment model based on the combination of backpropagation neural network (BPNN) and rough set theory (RST). First, a risk list with 35 risk factors were grouped into six risk categories via the brainstorming method and the original sample data set was constructed according to the initial risk list. Subsequently, an attribute reduction algorithm of the rough set was used to eliminate the redundancy attributes from the original sample dataset. The input factors of the software project risk assessment model could be reduced from thirty-five to twelve by the attribute reduction. Finally, the refined sample data subset was used to train the BPNN and the test sample data subset was used to verify the trained BPNN. The test results showed that the proposed joint model could achieve a better assessment than the model based only on the BPNN.

show abstract

Section: Data Preprocessing and Attribute Deductionmentioning

confidence: 99%