.png){kind=icon}
.png){kind=icon}
Return
Deep learning based QoE evaluation for internet video
.png){kind=icon}
PRE
AI
DOI：10.1016/j.neucom.2019.12.082.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
The issue of quality of experience (QoE) evaluation in Internet video is challenging for mainly two facts. One is the sophisticated interactions between features from different entities: user, system and context. The other is the heterogeneity of feature type, i.e., sequential and non-sequential characteristics of features. To address above challenges, we propose a hybrid network model that integrates a deep neural network (DNN) and an improved recurrent neural network (RNN) for representation learning of viewlevel QoE evaluation. Non-sequential side information and time difference of sequential features are incorporated to different layers of RNN. Attention mechanism is applied for further improvement of RNN. Based on the output of attention network, we propose a graph-based ranking algorithm to conduct user behavior analyzing, which is useful to online decision of service providers. We conduct experiments on a real-world dataset from a large-scale VoD services provider. The results demonstrate that our method is more effective over the state-of-the-art methods for QoE evaluation. (c) 2020 Elsevier B.V. All rights reserved.
Keywords:
QoE
Neural networks
User behavior
Time sequence
Graph-based ranking
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
6.5
Papers:
2.5W
Citations:
6.5W