Machine Learning Classifiers and Along with TPOT Classifier (Automl) to Predict the Readmission Patterns of Diabetic Patients
Phani Siginamsetty1, V. Krishna Reddy2
1Phani Siginamsetty, MTech Computer Science, KL University Vaddeswaram, Guntur District
2Dr.V. Krishna Reddy, Professor, KL University, Green Fields Vaddeswaram, Vijayawada
Manuscript received on April 02, 2020. | Revised Manuscript received on April 20, 2020. | Manuscript published on May 30, 2020. | PP: 688-695 | Volume-9 Issue-1, May 2020. | Retrieval Number: F7415038620/2020©BEIESP | DOI: 10.35940/ijrte.F7415.059120
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Diabetes is seen as a common problem in the present running world. And till date 470million people globally in 2019, and it might be increased to 676million by the end of 2045.So day to day the diabetic has become a major problem, and due to the current technologies, we can easily predict the readmission of a patient based upon his digenesis. In this paper we are using classification algorithms to solve the problem by early predictions. And we can check it by using multiple hybrid classifiers, whatever the algorithm gives the best accuracy we are considering it as the generic model and it is going to predict the future diabetic patients. And we are considering the diabetic dataset mainly it consists of multiple features based upon the data we will consider as independent and dependent data, and solve the problem. Here, in this paper the algorithms which we are going to use are Logistic Regression(LR), Decision Trees, Random Forest (RF),XGboost, Gaussian-Naïve Bayes, TPOT (automl).Out of them Random Forest gives the best accuracy which is about 95.2%, the accuracy is attained by following pre-processing stage in a good manner, and handled all missing data.
Keywords: Diabetes prediction, Logistic regression, decision trees, Random Forest, XGboost, Gaussian Naïve-Bayes, TPOT(AutoML)
Scope of the Article: Machine Learning