Towards Internet Scale Quality-of-Experience Measurement with Twitter

Abstract

Abstract. At present, Quality of Experience (QoE) measurements are accomplished by interrogating users for the perceived quality of a service they just have used. Influenced by many factors and often limited by domain or geographical region, this technique has several drawbacks when a general state of QoE for the internet as a whole is prospected. To achieve such a general metric, we leverage user complaints that we observe in real-time in social media. Such approaches have been successfully applied for the monitoring of specific and single services. We aim to extend existing methods in order to create an overall metric, define an internet wide QoE baseline, monitor changes and hence, provide a context for assessing smaller scale findings against a ground truth. The contribution of this work is to demonstrate the feasibility of using social media analysis for generating a meaningful value for quantifying the actual QoE of the internet.

Learning a QoE Metric from Social Media and Gaming Footage

“…(a) Complaints. We searched for the tags and terms shown in Table 1, which were partly inspired by prior work [15,16,19]. Each search was a combination of "#twitch" and one other tag/term, e.g., "#twitch #internetproblems" or "#twitch lag spikes."…”