Evaluating dimensionality reduction of comorbidities for predictive modeling in individuals with neurofibromatosis type 1
Dimensionality reduction through feature aggregation can enhance the performance of ML models, particularly in high-dimensional datasets with small sample sizes, commonly found in EHRs health applications. However, if not carefully optimized, it can lead to information loss and data oversimplification, potentially adversely affecting model performance.
Authors
- Randi E. Foraker (ORCID: https://orcid.org/0000-0001-9255-9394)
- Inez Y. Oh (ORCID: https://orcid.org/0000-0002-0780-8710)
- David H. Gutmann (ORCID: https://orcid.org/0000-0002-3127-5045)
- Philip Payne (ORCID: https://orcid.org/0000-0002-9532-2998)
- Aditi Gupta (ORCID: https://orcid.org/0000-0002-1807-8266)
- Stephanie M. Morris (ORCID: https://orcid.org/0000-0003-0461-1098)
- Zach Abrams
- Ethan Hillis
Institutions
- Kennedy Krieger Institute (US)
- Washington University in St. Louis (US)
- Center for Autism and Related Disorders (US)
Publication Details
- Journal
- JAMIA Open
- Published
- 2024-12-26
- DOI
- https://doi.org/10.1093/jamiaopen/ooae157
- Citations
- 1
- Primary Topic
- Neurofibromatosis and Schwannoma Cases
- Type
- article
- Field-Weighted Citation Impact
- 0.37
Funders
- National Institutes of Health
- National Institute of Neurological Disorders and Stroke