Author: Felipe Soares; Aline Villavicencio; Michel Jose Anzanello; Flavio Sanson Fogliatto; Marco Idiart; Mark Stevenson
Title: A novel high specificity COVID-19 screening method based on simple blood exams and artificial intelligence Document date: 2020_4_14
ID: 2s5xd1oc_40
Snippet: AI models may be affected by a specific partition of the data, that is, the particular samples used to calibrate and evaluate the model (i.e. training and test sets). To counter that, we repeated the process of training and evaluation 100 times using different partitions of data for training and test, and storing all information from each run. Approximately 90% of the data was assigned for training, and 10% for testing. Partitioning was carried o.....
Document: AI models may be affected by a specific partition of the data, that is, the particular samples used to calibrate and evaluate the model (i.e. training and test sets). To counter that, we repeated the process of training and evaluation 100 times using different partitions of data for training and test, and storing all information from each run. Approximately 90% of the data was assigned for training, and 10% for testing. Partitioning was carried out using stratified random sampling to ensure training and test sets with approximately the same proportion of positives and negatives.
Search related documents:
Co phrase search for related documents- AI model and negative positive: 1, 2, 3, 4, 5, 6
- AI model and specific partition: 1
- AI model and test set: 1, 2, 3, 4, 5, 6
- AI model and test training: 1, 2, 3
- evaluation training and negative positive: 1, 2, 3, 4
- evaluation training and test set: 1, 2, 3, 4, 5
- evaluation training and test set training: 1
- evaluation training and test training: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10
- negative positive and particular sample: 1, 2, 3
- negative positive and random sampling: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13
- negative positive and stratified random sampling: 1, 2, 3, 4
- negative positive and test set: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28
- negative positive and test set training: 1, 2, 3, 4
- negative positive and test training: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16
- negative positive and testing 10: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36
Co phrase search for related documents, hyperlinks ordered by date