Loading...

Application of Data Mining in Healtcare

Oliyaei, Azadeh | 2011

509 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 43102 (01)
  4. University: Sharif University of Technology
  5. Department: Industrial Engineernig
  6. Advisor(s): Salmasi, Nasser
  7. Abstract:
  8. Data mining is the one of top ten developing knowledge in the world. This study followed three fold objectives; Firstly, An efficient model based on data mining algorithms is proposed to predict the duration of hospitalization time for patients of digestive system disease that need short term care. Duration of hospitalization is an important criterion to be used for predicting the hospital resources. In order to, a combined model based on CHAID and C.5 decision trees and a neural network is suggested. The suggested model predict the duration of hospitalization with 82% accuracy. The second object of this study is to propose an algorithm based on likelihood ratio. The suggested algorithm calculates the possibility of problem target variable besides classifying the two-class problems. None of the algorithm in Data mining area could calculate the possibility of target variable. Using this algorithm, the mortality rate predicted with regarded to Co-Morbidity in hyperplasic left heart syndrome illness. This algorithm classifies the patients in to one of the group of patients who survive of this dieses and who die from this dieses with 90% accuracy. And the ultimate object is to propose an efficient model for multi-class classification problems by using the Support Vector Machines (SVM) algorithm. This algorithm has high efficiency to classify two-class. Developing an algorithm based on SVM in a multi-class problem in an efficient way is currently an open research area. The proposed algorithm, all classes are divided into two major classes based on the distance among center point of classes and range of spread data in those classes. According to the result the suggested algorithm provides more accuracy than the one-against-one, one-against-all and directed acyclic graph support vector machines algorithm
  9. Keywords:
  10. Data Mining ; Classification ; Likelihood Ratio ; Support Vector Machine (SVM) ; Health Care System ; Duration of Hospitalization

 Digital Object List

 Bookmark

No TOC