Predicting The Discharge of Patients Via Machine Learning Based Discharge Predictive Model

Abstract

The primary objective of the work is to create a discharge roster for patients by employing various machine learning techniques and to predict the discharge of a patient. The performance of the proposed discharge predictive model is measured through various performance measures. The research work is carried out based on the dataset formed with actual data of patients in hospital. The machine learning (ML) based Discharge Predictive Model is developed by combining well known ML algorithms like K-Nearest Neighbour (KNN) algorithm, Random Forests algorithm and Light Gradient Boosting algorithm with optimum parameters, various feature combinations and pre-processing techniques. The performance of the proposed model is measured in terms of accuracy and it is compared with various existing techniques like SVM and NN. The result of the comparison study exhibit that the proposed predictive learning model attained enhanced accuracy than other ML techniques.

Harnessing Data-Driven Insights: Predictive Modeling for Diamond Price Forecasting using Regression and Classification Techniques

“…While several studies have touched upon the domain of diamond price prediction, the majority have either focused on traditional statistical methods, rudimentary machine learning techniques. However, the true potential of advanced machine learning models [1], encompassing both regression and classification paradigms [2]- [5], remains largely unexplored in this context. The study is poised at this juncture, aiming to fill the gap in the literature.…”