Please use this identifier to cite or link to this item: http://hdl.handle.net/2080/4411
Full metadata record
DC FieldValueLanguage
dc.contributor.authorSabat, Naba Krushna-
dc.contributor.authorPati, Umesh Chandra-
dc.contributor.authorDas, Santos Kumar-
dc.date.accessioned2024-02-21T04:57:33Z-
dc.date.available2024-02-21T04:57:33Z-
dc.date.issued2024-02-
dc.identifier.citationNational conference on Intelligent Systems, IoT, and Wireless Communication for the Society (IIWCS), National Institute of Technology Rourkela 16-17 February 2024en_US
dc.identifier.urihttp://hdl.handle.net/2080/4411-
dc.descriptionCopyright belongs to proceeding publisheren_US
dc.description.abstractThe weather climate is contingent upon various meteorological factors, including humidity, temperature, and pressure, among others. The sudden alteration of these factors within the surrounding atmosphere begets detrimental consequences, spanning from the production sector to healthcare. As a result, weather prediction is required to prevent potentially hazardous situations with minimum loss. This paper presents an ensemble model that enhances the accuracy of predicting time-series weather data. The model has been examined using datasets from two different cities, namely Bengaluru and Dongsi. It has been observed that the model demonstrates superior accuracy in forecasting temperature fluctuations, seasonal patterns, and long-term climate variations. Furthermore, the model consistently outperforms several stateof-the-art models in terms of performance metrics such as MAE, MSE, RMSE, and R 2 Score.en_US
dc.subjectWeather predictionen_US
dc.subjectEnsemble modelen_US
dc.subjectTime-series analysisen_US
dc.subjectRoot Mean Squared Error (RMSE)en_US
dc.titleAn Ensemble Approach for Improving Time-series Weather Data Accuracyen_US
dc.typeArticleen_US
Appears in Collections:Conference Papers

Files in This Item:
File Description SizeFormat 
2024_IIWCS_NKSabat_AnEnsemble.pdf620.94 kBAdobe PDFView/Open    Request a copy


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.