Manual Feature Engineering Task
(Redirected from man-made input feature)
Jump to navigation
Jump to search
A Manual Feature Engineering Task is a feature engineering task that is a manual software engineering task.
- Context:
- It can involve ML Feature Development Task, ML Feature Monitoring Task, ML Feature Maintenance Task, ...
- …
- Example(s):
- add customer's monthly visit count to our feature space.
- analyze why the customer's monthly visit count feature has dramatically changed its distribution in the past day.
- repair a data error in the customer's monthly visit count feature.
- …
- Counter-Example(s):
- See: Feature Extraction.
References
2017
- https://burakhimmetoglu.com/2017/08/22/time-series-classification-with-tensorflow/
- QUOTE: … A standard approach to time-series problems usually requires manual engineering of features which can then be fed into a machine learning algorithm. Engineering of features generally requires some domain knowledge of the discipline where the data has originated from. For example, if one is dealing with signals (i.e. classification of EEG signals), then possible features would involve power spectra at various frequency bands, Hjorth parameters and several other specialized statistical properties.
2011
- (Collobert et al., 2011) ⇒ Ronan Collobert, Jason Weston, Léon Bottou, Michael Karlen, Koray Kavukcuoglu, and Pavel Kuksa. (2011). “Natural Language Processing (Almost) from Scratch.” In: The Journal of Machine Learning Research, 12.
- QUOTE: Instead of exploiting man-made input features carefully optimized for each task, our system learns internal representations on the basis of vast amounts of mostly unlabeled training data.