|
Selection of Variables by the F-Score Algorithm for Radiated Magnetic Field Signals Discrimination of Electrical DischargesEasyChair Preprint 359, version historyVersion | Date | Pages | Version notes |
---|
1 | July 20, 2018 | 5 | | 2 | October 7, 2018 | 5 | We prefer to replace the term learning with the term training, much more used in machine learning. We have summarized the F-Score algorithm with the proposed threshold, in the form of a flowchart as proposed by one of the reviewers. We also refined and expanded the values of the C and Sigma parameters of the model, so that we found new values of the DSSAT recognition rate equal to 33.33% instead of 28.57% and therefore the training and the testing times bigger. We have corrected some references, in particular in reference 8 where we replaced "regression" instead of regesssion and add "page to page" to references 8 and 9. |
Keyphrases: Electrical discharges, F-score, Insulation Systems, Support Vector Machines (SVMs), Variables Selection |
|
|