Data Validation Process in Machine Learning Pipeline |
Author(s): |
| Ram Mohan Vadavalasa |
Keywords: |
| Machine Learning Pipeline, Data Validation Process |
Abstract |
|
The Machine learning is a powerful tool for finding patterns from massive amounts of data. Data is being generated, collected, transformed, processed, and analyzed for machine learning end to end life cycle. Machine learning research has focused on improving the accuracy and efficiency of training algorithms, but there is an equally important problem of monitoring the quality of data fed into machine learning. Machine learning pipeline treats training and serving data as an important product asset, which is equal to the algorithm and infrastructure used for learning. Validating data is an essential requirement to certify the worthiness and benchmark of the Machine Learning system. |
Other Details |
|
Paper ID: IJSRDV8I40680 Published in: Volume : 8, Issue : 4 Publication Date: 01/07/2020 Page(s): 449-452 |
Article Preview |
|
|
|
|
