Predicting QoE Factors with Machine Learning

Vasilev, Vladislav; Leguay, Jérémie; Paris, Stefano; Maggi, Lorenzo; Debbah, Mérouane

doi:10.1109/icc.2018.8422609

Cited by 35 publications

(38 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…While we also use supervised ML in our work, our focus is not on the final QoE models, but on reducing the cost of building such models by active sampling. Furthermore, the conventional approach of using supervised ML in QoS-QoE modeling is to use the training data that is either collected in the wild [31], [7] or built by controlled experimentation [32]. For data collected in the wild, the model suffers from the biasness of the underlying data source, while for the case of controlled experimentation, the similarity in the experimental space is not fully exploited.…”

Section: Related Work and Discussionmentioning

confidence: 99%

An intelligent sampling framework for controlled experimentation and QoE modeling

Khokhar¹,

Saber²,

Spetebroot³

et al. 2018

Computer Networks

View full text Add to dashboard Cite

For internet applications, measuring, modeling and predicting the quality experienced by end users as a function of network conditions is challenging. A common approach for building application specific Quality of Experience (QoE) models is to rely on controlled experimentation. For accurate QoE modeling, this approach can result in a large number of experiments to carry out because of the multiplicity of the network features, their large span (e.g., bandwidth, delay) and the time needed to setup the experiments themselves. However, most often, the space of network features in which experimentations are carried out shows a high degree of similarity in the training labels of QoE. This similarity, difficult to predict beforehand, amplifies the training cost with little or no improvement in QoE modeling accuracy. So, in this paper, we aim to exploit this similarity, and propose a methodology based on active learning, to sample the experimental space intelligently, so that the training cost of experimentation is reduced. We validate our approach for the case of YouTube video streaming QoE modeling from out-of-band network performance measurements, and perform a rigorous analysis of our approach to quantify the gain of active sampling over uniform sampling.

show abstract

Section: Related Work and Discussionmentioning

confidence: 99%

An intelligent sampling framework for controlled experimentation and QoE modeling

Khokhar¹,

Saber²,

Spetebroot³

et al. 2018

Computer Networks

View full text Add to dashboard Cite

show abstract

“…An example is [14], which proposed a proactive service quality assessment process called Q-Score to estimate a single QoE metric using the network performance metrics. An ML-based QoE prediction approach is introduced in [15] that can be applied in a proactive approach. It predicts rebuffering events from QoS parameters using a Bayesian Network; hence, it needs to monitor network parameters.…”

Section: Related Workmentioning

confidence: 99%

Forecasting Video QoE With Deep Learning From Multivariate Time-Series

Dinaki

Shirmohammadi

Janulewicz

et al. 2021

IEEE Open J. Signal Process.

View full text Add to dashboard Cite

The end users' satisfactory Quality of Experience (QoE) is a fundamental criterion for networked video service providers such as video-on-demand providers (Netflix, YouTube, etc.), cloud gaming providers (Google Stadia, PlayStation Now, etc.) and videoconferencing providers (Zoom, Microsoft Teams, etc.). To know the QoE, providers today typically predict it from the Quality of Service (QoS) parameters or the client-side's actual QoE metrics measured at the current time-step. But the former does not precisely reflect the users' experience, and the latter has a delay between QoE measurements at the client-side and the user's current experience. Mitigating this delay can provide a noticeable improvement in the delivery system's performance. For example, accurate forecasting of QoE for the near future allows the service management system to take a proactive approach and fix delivery issues before they become a noticeable problem at the end user, or at least reduce overall QoE degradation. QoE forecasting can also be used in rate adaptation in DASH or resource allocation in wireless networks. In this paper, we propose a method to prognosticate QoE metrics. Using data collected from an industry video streaming testbed for three different classes, we define a multivariate time series forecasting problem. We then model a hybrid state-of-the-art deep learning method, BiLSTM-CNN, to forecast the QoE metrics in advance. Evaluation of our proposed method compared to four other well-known ML models of Support Vector Regression (SVR), Multi-Layer Perceptron (MLP), Long short-term memory (LSTM), and Bidirectional LSTM (BiLSTM) demonstrates the superior performance of our proposed method.

show abstract

“…They also infer that viewers tend to watch video for lesser duration as they experience more rebuffering due to video stalls. Vasilev et al [6] propose a Bayesian network model to predict re-buffering ratio or the Stall variable and try to improve the accuracy of prediction through Logistic regression. Zanforlin et al [10] expressed video QoE in terms of the average structural similarity or SSIM index and use this to tag videos with polynomial co-efficients that provide a compact description of its specific SSIM behavior.…”

Section: Related Workmentioning

confidence: 99%

A Learning Framework to Improve Video QoE in HTTP Adaptive Streaming SDN for Delivering Live Content

Natarajan

Krishnaveni

2019

International Journal of Simulation: Systems, Science &Amp; Technology

View full text Add to dashboard Cite

HTTP (Hyper Text Transport Protocol) adaptive streaming techniques suffers from start-up latencies and video stalls which impair Video Quality of Experience (QoE). We deal with these issues in 4 steps: i) we analyze the advantages of variable bit rate encoded streams on Adaptive Bit Rate (ABR) streaming and assess the impact of chunk sizes on video quality and scene complexity; ii) we propose a novel machine learning framework through a Gaussian Mixture model k-means clustering to isolate video stall events, the Machine Learning Pipeline runs these Support Vector Machines (SVM) algorithms to accurately predict the occurrence and duration of video stalls; iii) we employ Software Defined Networking (SDN) based OpenFlow controller to dynamically switch, re-prioritize tracks and chunk sizes across tracks to improve the overall Video QoE; iv) we further optimize this through an intelligent prediction of video stall occurrence and stall duration in accordance with network traffic conditions and scene complexity without compromising video quality.

show abstract

Predicting QoE Factors with Machine Learning

Cited by 35 publications

References 16 publications

An intelligent sampling framework for controlled experimentation and QoE modeling

An intelligent sampling framework for controlled experimentation and QoE modeling

Forecasting Video QoE With Deep Learning From Multivariate Time-Series

A Learning Framework to Improve Video QoE in HTTP Adaptive Streaming SDN for Delivering Live Content

Contact Info

Product

Resources

About