An empirical evaluation of dimensionality reduction and class balancing for medical text classification
Arslan Jamil, Muhammad Kashif Hanif, Muhammad Umer Sarwar, Muhammad Irfan Khan · 2025
Read An empirical evaluation of dimensionality reduction and class balancing for medical text classification by Arslan Jamil, Muhammad Kashif Hanif, Muhammad Umer Sarwar, Muhammad Irfan Khan (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-12780033.txt.
Description
Arslan Jamil and three colleagues published a computing study in two thousand twenty-five on classifying medical text efficiently. Clinical notes are long and sparse, which slows learning models. Shrinking the feature space and balancing rare classes may fix both problems.
The team paired three shrinking methods with a synthetic oversampling trick for rare classes. The classic combination of principal component analysis plus oversampling won. It reached about ninety-one percent accuracy on a standard sample corpus. It also improved a fairness-style score over the raw baseline and trained about forty-two percent faster. The authors offer the pipeline as a practical route for hospitals handling large note archives.
Catalog record
| CREATOR | Arslan Jamil, Muhammad Kashif Hanif, Muhammad Umer Sarwar, Muhammad Irfan Khan |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 117 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |