Please use this identifier to cite or link to this item:
Title: An Ensemble Feature Selection Method for Prediction of Chronic Diseases
Researcher: Manonmani M
Guide(s): Sarojini B
Keywords: Engineering and Technology
Computer Science
Computer Science Interdisciplinary Applications
University: Avinashilingam Institute for Home Science and Higher Education for Women
Completed Date: 2021
Abstract: As large and complex datasets are becoming increasingly available to the research community, newlinemore advanced and sophisticated data analytical techniques are needed to exploit and manage these newlinedata. Machine learning and data mining methods can be used to mine significant knowledge from a newlinevariety of large and heterogeneous data sources, supporting biomedical research and healthcare newlinedelivery. Data-driven healthcare is at the center of the vision of learning health systems and holds newlinegreat promise for transforming the current healthcare status. Predictive data mining is becoming an newlineimportant analytical instrument for the scientific community and clinical practitioners in the field of newlinemedicine. Secondary use of patient and clinical study data is able to enhance health care experiences newlinefor individuals. Further, it enables the expansion of knowledge about diseases and treatments and leads newlineto an increase in the efficiency and effectiveness of health care systems. newlineThe proposed research work aims at developing a prediction and classification model known as newlineD-ITLBO that selects relevant features in medical datasets based on an ensemble feature selection newlineprocess. In the first stage of feature selection, a filter-based approach known as Density based Feature newlineSelection (DFS) method is applied to five chronic medical datasets. The DFS method ranks features newlinebased on the Probability Density Function (PDF) and selects the relevant features. In the next stage, newlinethe derived feature subset from DFS method is given as input to a wrapper- based optimized feature newlineselection algorithm known as Improved Teacher Learner Based Optimization (ITLBO) Algorithm. newlineThe ITLBO algorithm combined with Recursive Feature Elimination (RFE) technique transforms the newlinehigh dimensional dataset to a low dimensional solution space and provides a set of most newlinediscriminatory features that are almost as informative as the original dataset for prediction of chronic newlinediseases. The proposed ensemble feature selection method, D-ITLBO is an ensemble of the filterbased newlineDFS m
Pagination: 196 p.
Appears in Departments:Department of Computer Science

Files in This Item:
File Description SizeFormat 
01_title.pdfAttached File147.57 kBAdobe PDFView/Open
02_certificate.pdf162.21 kBAdobe PDFView/Open
03_acknowledgement.pdf154.66 kBAdobe PDFView/Open
04_contents.pdf171.67 kBAdobe PDFView/Open
05_list of tables, figures and abbreviations.pdf290.36 kBAdobe PDFView/Open
06_chapter 1.pdf419.36 kBAdobe PDFView/Open
07_chapter 2.pdf404.57 kBAdobe PDFView/Open
08_chapter 3.pdf458.38 kBAdobe PDFView/Open
09_chapter 4.pdf456.04 kBAdobe PDFView/Open
10_chapter 5.pdf964.9 kBAdobe PDFView/Open
11_chapter 6.pdf1.63 MBAdobe PDFView/Open
12_chapter 7.pdf346.44 kBAdobe PDFView/Open
13_chapter 8.pdf161.66 kBAdobe PDFView/Open
14_references.pdf399.81 kBAdobe PDFView/Open
80_recommendation.pdf146.95 kBAdobe PDFView/Open
Show full item record

Items in Shodhganga are licensed under Creative Commons Licence Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0).

Altmetric Badge: