Machine learning techniques for predicting hospital length of stay in pennsylvania federal and specialty hospitals

Parag C. Pendharkar, Hitesh Khurana

Research output: Contribution to journalArticle

6 Citations (Scopus)

Abstract

In this paper, we compare three different machine learning techniques for predicting length of stay (LOS) in Pennsylvania Federal and Specialty hospitals. Using the real-world data on 88 hospitals, we compare the performances of three different machine learning techniques-Classification and Regression Tree (CART), Chi-Square Automatic Interaction Detection (CHAID) and Support Vector Regression (SVR)-and find that there is no significant difference in performances of these three techniques. However, CART provides a decision tree that is easy to understand and interpret. The results from CART indicate that psychiatric care hospitals typically have higher LOS than nonpsychiatric care hospitals. For non-psychiatric care hospitals, the LOS depends on hospital capacity (beds staffed) with larger hospitals with beds staffed over 329 having average LOS of 13 weeks vs. smaller hospitals with average LOS of about 3 weeks.

Original languageEnglish (US)
Pages (from-to)45-56
Number of pages12
JournalInternational Journal of Computer Science and Applications
Volume11
Issue number3
StatePublished - Jan 1 2014

Fingerprint

Learning systems
Hospital beds
Decision trees

All Science Journal Classification (ASJC) codes

  • Computer Science Applications

Cite this

@article{049b6d0fcc2149ef9a1146a053c30b7f,
title = "Machine learning techniques for predicting hospital length of stay in pennsylvania federal and specialty hospitals",
abstract = "In this paper, we compare three different machine learning techniques for predicting length of stay (LOS) in Pennsylvania Federal and Specialty hospitals. Using the real-world data on 88 hospitals, we compare the performances of three different machine learning techniques-Classification and Regression Tree (CART), Chi-Square Automatic Interaction Detection (CHAID) and Support Vector Regression (SVR)-and find that there is no significant difference in performances of these three techniques. However, CART provides a decision tree that is easy to understand and interpret. The results from CART indicate that psychiatric care hospitals typically have higher LOS than nonpsychiatric care hospitals. For non-psychiatric care hospitals, the LOS depends on hospital capacity (beds staffed) with larger hospitals with beds staffed over 329 having average LOS of 13 weeks vs. smaller hospitals with average LOS of about 3 weeks.",
author = "Pendharkar, {Parag C.} and Hitesh Khurana",
year = "2014",
month = "1",
day = "1",
language = "English (US)",
volume = "11",
pages = "45--56",
journal = "International Journal of Computer Science and Applications",
issn = "0972-9038",
publisher = "Technomathematics Research Foundation",
number = "3",

}

TY - JOUR

T1 - Machine learning techniques for predicting hospital length of stay in pennsylvania federal and specialty hospitals

AU - Pendharkar, Parag C.

AU - Khurana, Hitesh

PY - 2014/1/1

Y1 - 2014/1/1

N2 - In this paper, we compare three different machine learning techniques for predicting length of stay (LOS) in Pennsylvania Federal and Specialty hospitals. Using the real-world data on 88 hospitals, we compare the performances of three different machine learning techniques-Classification and Regression Tree (CART), Chi-Square Automatic Interaction Detection (CHAID) and Support Vector Regression (SVR)-and find that there is no significant difference in performances of these three techniques. However, CART provides a decision tree that is easy to understand and interpret. The results from CART indicate that psychiatric care hospitals typically have higher LOS than nonpsychiatric care hospitals. For non-psychiatric care hospitals, the LOS depends on hospital capacity (beds staffed) with larger hospitals with beds staffed over 329 having average LOS of 13 weeks vs. smaller hospitals with average LOS of about 3 weeks.

AB - In this paper, we compare three different machine learning techniques for predicting length of stay (LOS) in Pennsylvania Federal and Specialty hospitals. Using the real-world data on 88 hospitals, we compare the performances of three different machine learning techniques-Classification and Regression Tree (CART), Chi-Square Automatic Interaction Detection (CHAID) and Support Vector Regression (SVR)-and find that there is no significant difference in performances of these three techniques. However, CART provides a decision tree that is easy to understand and interpret. The results from CART indicate that psychiatric care hospitals typically have higher LOS than nonpsychiatric care hospitals. For non-psychiatric care hospitals, the LOS depends on hospital capacity (beds staffed) with larger hospitals with beds staffed over 329 having average LOS of 13 weeks vs. smaller hospitals with average LOS of about 3 weeks.

UR - http://www.scopus.com/inward/record.url?scp=84923014193&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84923014193&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:84923014193

VL - 11

SP - 45

EP - 56

JO - International Journal of Computer Science and Applications

JF - International Journal of Computer Science and Applications

SN - 0972-9038

IS - 3

ER -