Implementation of Quality of Experience Prediction Framework through Mobile Network Data
Ayisat W. Yusuf-Asaju1, Zulkhairi MD. Dahalin2, Azman Taa3
1Ayisat W. Yusuf-Asaju, Department of Computer Science, University of Ilorin, Para Medical Board, Kwara-State Nigeria, School of Computing, University Utara Malaysia, Sintok, Malaysia.
2Zulkhairi MD. Dahalin, School of Computing, University Utara Malaysia, Sintok, Malaysia.
3Azman Taa, School of Computing, University Utara Malaysia, Sintok, Malaysia.
Manuscript received on 10 June 2019 | Revised Manuscript received on 17 June 2019 | Manuscript Published on 19 June 2019 | PP: 693-700 | Volume-8 Issue-8S June 2019 | Retrieval Number: H11190688S19/19©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open-access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Generally, a reliable method of analysing the quality of experience is through the subjective method, which is time consuming, lacks usability, lacks repeatability in real-time and near real-time. Another method is the objective measurement that aims at predicting the subjective measurement based on the estimated mean opinion score. Therefore, this study adopted the objective measurement by implementing a quality of experience framework, which employed predictive analytics techniques to analyse the mobile internet user experience dataset gathered through the mobile network. The predictive analytics employed the use of multiple regression, neural network, decision trees, random forest, and decision forest to predict the mobile internet perceived quality of experience. Result from the study shows that decision forests perform better than other algorithms used for the predictive analytics. In addition, the result indicates that the predictive analytics can be used to enhance the allocation of network resources based on location and time constituted in the dataset.
Keywords: Internet Service, Subscribers, Prediction, Real-Time, Machine Learning.
Scope of the Article: Machine Learning