Please use this identifier to cite or link to this item:
https://hdl.handle.net/20.500.11851/11776
Title: | XGBoost-based QoE Prediction for Mobile Networks | Other Titles: | Mobil Ağlar için XGBoost Tabanlı QoE Tahmini | Authors: | Dayı, A.B. Tuna, E. |
Keywords: | 6G Decision Tree Multimedia systems Quality of experience Quality of service Wireless communication XGBoost Mobile telecommunication systems Multimedia systems Quality of service 6g Complexes structure Network applications Quality of experience Quality-of-service Users perspective Users' experiences Users' satisfactions Wireless communications Xgboost Decision trees |
Publisher: | Institute of Electrical and Electronics Engineers Inc. | Abstract: | The increasing demand for advanced video-based services necessitates operators to ensure the most suitable network performance while also considering user satisfaction with the service. QoS provides significant insights on the network side to deliver satisfactory user experiences. On the other hand, QoE informs about how a given service is perceived from the user's perspective. The more advanced video-based services to be offered with the more complex structure of 6G increase the importance of mapping QoS to QoE. This paper presents an XGBoost-based method for predicting QoE based on UE-based, network-based, and application-based QoS measurements obtained from a real and live mobile network. The results indicate that XGBoost is an effective method for user experience estimation. © 2024 IEEE. | Description: | Berdan Civata B.C.; et al.; Figes; Koluman; Loodos; Tarsus University 32nd IEEE Conference on Signal Processing and Communications Applications, SIU 2024 -- 15 May 2024 through 18 May 2024 -- Mersin -- 201235 |
URI: | https://doi.org/10.1109/SIU61531.2024.10600936 https://hdl.handle.net/20.500.11851/11776 |
ISBN: | 979-835038896-1 |
Appears in Collections: | Scopus İndeksli Yayınlar Koleksiyonu / Scopus Indexed Publications Collection |
Show full item record
CORE Recommender
Items in GCRIS Repository are protected by copyright, with all rights reserved, unless otherwise indicated.