| A LITERATURE REVIEW OF POPULAR TIME SERIES ANALYSIS AND FORECASTING METHODS |
| Paper ID : 1115-ICEEM2023 (R1) |
| Authors |
|
Mohamed khalil Elnagar *1, Taha E.Taha2, Fathi Fathi E. Abd El‑Samie2, adel elfishawy2 1Menofia university .Electronics and Electrical Communications Engineering, Menouf
Menouf, Egypt 2Menofia university .Electronics and Electrical Communications Engineering, Menouf |
| Abstract |
| Abstract— Time series are widely utilized in various fields, serving as the foundation for applications such as economic forecasting, sales projections, solar wind analysis, stock market evaluation, yield estimations, health monitoring and disease prediction. For specific medical research purposes, electronic health records containing patient diagnostic information provide valuable indications in analyzing and extracting insights from these time-based EHR datasets. Time-series analysis techniques employed to enhance heart failure prediction through robust architecture development. In terms of methodology for analyzing time series data across different domains and industries including healthcare, it includes traditional linear modeling approaches like Statistical models such as ARIMA or traditional linear machine learning algorithms; furthermore, deep learning models - RNN LSTM CNN; along with automated machine-learning frameworks Auto ML. |
| Keywords |
| time series analysis, machine learning, deep learning, electronic health records (EHRs) |
| Status: Accepted |