L(a)ying in (Test)Bed - How Biased Datasets Produce Impractical Results for Actual Malware Families' Classification.
Tamy BepplerMarcus BotacinFabricio CeschinLuiz E. S. OliveiraAndré GrégioPublished in: ISC (2019)
Keyphrases
- test bed
- computer vision systems
- test beds
- uci repository
- benchmark datasets
- uci machine learning repository
- classification scheme
- machine learning
- pattern recognition
- preprocessing
- support vector machine svm
- neural network
- malware detection
- uci datasets
- classification systems
- classification method
- computer vision
- feature vectors
- classification accuracy
- binary and multi class
- space station
- data sets
- feature extraction
- automatic classification
- support vector machine
- classification algorithm
- cost sensitive
- pattern classification
- training set
- feature space
- support vector
- resource allocation
- decision trees
- multi class
- text classification
- model selection