On the need of preserving order of data when validating within-project defect classifiers.
Davide FalessiJacky HuangLikhita NarayanaJennifer Fong ThaiBurak TurhanPublished in: Empir. Softw. Eng. (2020)
Keyphrases
- data sets
- data collection
- raw data
- data sources
- prior knowledge
- high quality
- training data
- data structure
- original data
- small number
- synthetic data
- data points
- image data
- sensor data
- data quality
- data samples
- statistical methods
- database
- test data
- experimental data
- high dimensional data
- statistical analysis
- multi class
- query processing
- decision trees
- metadata
- neural network