Context Modeling in Predictive Analytics
Anton Alekseevich Romanov, Aleksey A. Filippov · 2021 International Conference on Information Technology and Nanotechnology (ITNT) · 2021
The state forecasting of complex systems requires increasing the reliability of modeling results. The creation of a hybrid predictive model is needed to solve this task. The contextual information extracted from the subject area in data analysis may improve the quality of time series forecasts. The proposition is in using an ontology for the representation of time series context. The logical model of the ontology in descriptive logic ALCHI (D) and the automated approach to its formation are considered. A comparison of time series forecasting methods is presented.