| ||||
| ||||
![]() Title:Evaluation of Classical and AI-Based Time Series Models for Health Forecasting with Algorithms Conference:TMREES26Lb Tags:Artificial Intelligence, Forecasting, Healthcare, Nonlinear models, SVM and Traditional Abstract: In this research, alternative models are proposed for predicting time series of health phenomena using some artificial intelligence techniques to increase accuracy and improve estimation. These are based on artificial intelligence methods, namely support vector machines (SVM) and artificial neural networks (NNs) to deal with nonlinear phenomena, in addition to estimating the well-known traditional models represented by the autoregressive model (AR), the moving mean model (MA), the autoregressive-moving mean model (ARMA), the autoregressive-integral model (ARIMA), and the autoregressive model with exogenous variables (ARX). Simulation experiments were conducted on experimental data, and a real data set represented by the number of cancer cases was applied. It was found that traditional models perform predictions quickly but with less accuracy, while artificial intelligence-based models are characterized by high accuracy and speed. Evaluation of Classical and AI-Based Time Series Models for Health Forecasting with Algorithms ![]() Evaluation of Classical and AI-Based Time Series Models for Health Forecasting with Algorithms | ||||
| Copyright © 2002 – 2026 EasyChair |
