{
  "id": "1810.09942",
  "version": "v1",
  "published": "2018-10-23T16:14:11.000Z",
  "updated": "2018-10-23T16:14:11.000Z",
  "title": "Preprocessor Selection for Machine Learning Pipelines",
  "authors": [
    "Brandon Schoenfeld",
    "Christophe Giraud-Carrier",
    "Mason Poggemann",
    "Jarom Christensen",
    "Kevin Seppi"
  ],
  "comment": "Accepted at the ICML 2018 AutoML Workshop",
  "categories": [
    "cs.LG",
    "stat.ML"
  ],
  "abstract": "Much of the work in metalearning has focused on classifier selection, combined more recently with hyperparameter optimization, with little concern for data preprocessing. Yet, it is generally well accepted that machine learning applications require not only model building, but also data preprocessing. In other words, practical solutions consist of pipelines of machine learning operators rather than single algorithms. Interestingly, our experiments suggest that, on average, data preprocessing hinders accuracy, while the best performing pipelines do actually make use of preprocessors. Here, we conduct an extensive empirical study over a wide range of learning algorithms and preprocessors, and use metalearning to determine when one should make use of preprocessors in ML pipeline design.",
  "revisions": [
    {
      "version": "v1",
      "updated": "2018-10-23T16:14:11.000Z"
    }
  ],
  "analyses": {
    "keywords": [
      "machine learning pipelines",
      "preprocessor selection",
      "data preprocessing hinders accuracy",
      "ml pipeline design",
      "solutions consist"
    ],
    "note": {
      "typesetting": "TeX",
      "pages": 0,
      "language": "en",
      "license": "arXiv",
      "status": "editable"
    }
  }
}