Requirements-Driven Method to Determine Quality Characteristics and Measurements for Machine Learning Software and Its Evaluation.
Koji NakamichiKyoko OhashiIsao NambaRieko YamamotoMikio AoyamaLisa JöckelJulien SiebertJens HeidrichPublished in: RE (2020)
Keyphrases
- machine learning
- machine learning methods
- high precision
- evaluation method
- gold standard
- preprocessing
- pairwise
- significant improvement
- high quality
- prior knowledge
- high accuracy
- software development
- case study
- clustering method
- evaluation model
- quality assessment
- data sets
- classification method
- machine learning algorithms
- detection method
- unsupervised learning
- feature set
- edge detection
- classification accuracy
- cost function
- computational complexity