Closed lucas-lr closed 6 years ago
We select a commun amount of features that we will request to our clients. Our algorithm will enreach them automatically taking in account their meaning. Extra features may be also included but the algorithm won't take advantatge of the domain when cleaning and enriching them.
Descripcion of commun features on the wiki: https://github.com/AgileDataScienceUB/ADS4/wiki/Data-consolidation
We have to select the best features from the datasets. Indeed, some of them are more useful than others, so we should focus on them, ignoring the others.
Estimated time: 8 hours