Web browsing is a very common way of using the Internet to, among others, read news, do on-line shopping, or search for user generated content such as YouTube or Dailymotion. Traditional evaluations of web surfing focus on objectively measured Quality of Service (QoS) metrics such as loss rate or round-trip times; however, little is known how these QoS metrics relate to the user satisfaction, referred to as Quality of Experience (QoE). In this paper, we propose to use K-means clustering to discover the relationship between the subjective QoE and the objective QoS: Each Web session is described by a so called ‘signature’ that consists of a set QoS metrics and the number of elements the Web page is composed of. In addition, we use a browser plugin to measure the time it takes to render the entire Web page (full load time) and ask the user to express via a feedback button its (dis-)satisfaction with the speed at which the Web page was rendered. Clustering the Web sessions of multiple users based on their signatures allows to discover and explain the performance differences among users and identify the relationship between the QoS measured and the QoE experienced: User dis-satisfaction is often related to large round-trip delays, high loss rates, or Web pages with a large number of elements. We also see that there is a strong correlation between the full load time and the QoE: a full load time of ten seconds or more is typically not acceptable for the users. Index Terms Web Browsing, Home Networks Measurement, Quality of Experiences, Quality of Service
[1]
Jaideep Chandrashekar,et al.
HostView: annotating end-host performance measurements with user feedback
,
2011,
SIGMETRICS Perform. Evaluation Rev..
[2]
Heng Cui,et al.
Trouble shooting interactive web sessions in a home environment
,
2011,
HomeNets '11.
[3]
Kuan-Ta Chen,et al.
OneClick: A Framework for Measuring Network Quality of Experience
,
2009,
IEEE INFOCOM 2009.
[4]
Jaideep Chandrashekar,et al.
Performance of networked applications: the challenges in capturing the user's perception
,
2011,
W-MUST '11.
[5]
Matti Siekkinen,et al.
A root cause analysis toolkit for TCP
,
2008,
Comput. Networks.
[6]
Andrew W. Moore,et al.
X-means: Extending K-means with Efficient Estimation of the Number of Clusters
,
2000,
ICML.
[7]
Aymen Hafsaoui,et al.
Application Level Performance in Wired and Wireless Environments. (Peformances de Niveau Applicatif en Environnement Filaire et sans Fil)
,
2011
.
[8]
Marco Mellia,et al.
Mining Unclassified Traffic Using Automatic Clustering Techniques
,
2011,
TMA.
[9]
J. Pearl.
Causality: Models, Reasoning and Inference
,
2000
.
[10]
Matti Siekkinen,et al.
Understanding the Impact of the Access Technology: The Case of Web Search Services
,
2011,
TMA.