Symmetry Based Feature Selection with Multi layer Perceptron for the prediction of Chronic Disease

Abstract : Huge amount of Healthcare data are produced every day from the various health care sectors. The accumulated data can be effectively analyzed to identify people's risk from chronic diseases. The process of predicting the presence or absence of the disease and also to diagnosing the various disease using the historical medical data is known as Health Care Analytics. Health care analytics will improve patient care and also the harness practice of medical practitioner. The feature selection is considered as a core aspect of the machine learning which hugely contribute towards the performance of the machine learning model. In this paper symmetry based feature subset selection is proposed to select the optimal features from the Health care data which contribute towards the prediction outcome. The Multilayer perceptron algorithm(MLP) used as a classifier which will predict the outcome by using the features which are selected from the Symmetry-based feature subset selection technique. The chronic disease dataset Diabetes, Cancer, Breast Cancer, and Heart Disease data set accumulated from UCI repository is used to conduct the experiment. The experimental results demonstrate that the proposed hybrid combination of feature selection technique and the multilayer perceptron outperforms in accuracy compare to the existing approaches.
Complete list of metadatas

Cited literature [16 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-02265618
Contributor : Galgotias Singh <>
Submitted on : Saturday, August 10, 2019 - 8:42:33 PM
Last modification on : Monday, August 12, 2019 - 12:29:47 PM

File

B2658078219.pdf
Publisher files allowed on an open archive

Identifiers

Citation

Sandeepkumar Hegde, Rajalaxmi Hedge. Symmetry Based Feature Selection with Multi layer Perceptron for the prediction of Chronic Disease. International Journal of Recent Technology and Engineering, Blue Eyes Intelligence Engineering & Sciences Publication, 2019, 8 (2), pp.3316-3322. ⟨https://www.ijrte.org/wp-content/uploads/papers/v8i2/B2658078219.pdf⟩. ⟨10.35940/ijrte.B2658.078219⟩. ⟨hal-02265618⟩

Share

Metrics

Record views

14

Files downloads

8