{
  "id": "2307.02071",
  "version": "v1",
  "published": "2023-07-05T07:26:27.000Z",
  "updated": "2023-07-05T07:26:27.000Z",
  "title": "A Comparison of Machine Learning Methods for Data with High-Cardinality Categorical Variables",
  "authors": [
    "Fabio Sigrist"
  ],
  "categories": [
    "cs.LG",
    "cs.AI",
    "stat.ML"
  ],
  "abstract": "High-cardinality categorical variables are variables for which the number of different levels is large relative to the sample size of a data set, or in other words, there are few data points per level. Machine learning methods can have difficulties with high-cardinality variables. In this article, we empirically compare several versions of two of the most successful machine learning methods, tree-boosting and deep neural networks, and linear mixed effects models using multiple tabular data sets with high-cardinality categorical variables. We find that, first, machine learning models with random effects have higher prediction accuracy than their classical counterparts without random effects, and, second, tree-boosting with random effects outperforms deep neural networks with random effects.",
  "revisions": [
    {
      "version": "v1",
      "updated": "2023-07-05T07:26:27.000Z"
    }
  ],
  "analyses": {
    "keywords": [
      "machine learning methods",
      "high-cardinality categorical variables",
      "random effects outperforms deep neural",
      "effects outperforms deep neural networks"
    ],
    "note": {
      "typesetting": "TeX",
      "pages": 0,
      "language": "en",
      "license": "arXiv",
      "status": "editable"
    }
  }
}