AT&T Labs Research — Leading Invention, Driving Innovation

{Understanding mobile customer experience and behavior is an important task for cellular service providers to improve the satisfaction of their customers. To that end, cellular service providers regularly measure the properties of their mobile network, such as signal strength, dropped calls, call blockage, and radio interface failures (RIFs). In addition to these passive measurements collected within the network, understanding customer sentiment from direct customer feedback is also an important means of evaluating user experience. Customers have varied perceptions of mobile network quality, and also react differently to advertising, news articles, and the introduction of new equipment and services. Traditional methods used to assess customer sentiment include direct surveys and mining the transcripts of calls made to customer care centers. Along with this feedback provided directly to the service providers, the rise in social media potentially presents new opportunities to gain further insight into customers by mining public social media data as well. According to a note from one of the largest online social network (OSN) sites in the US [7], as of September 2010 there are 175 million registered users, and 95 million text messages communicated among users per day. Additionally, many OSNs provide APIs to retrieve publically available message data, which can be used to collect this data for analysis and interpretation. Our plan is to correlate different sources of measurements and user feedback to understand the social media usage patterns from mobile data users in a large nationwide cellular network. In particular, we are interested in quantifying the traffic volume, the growing trend of social media usage and how it interacts with traditional communication channels, such as voice calls, text messaging, etc. In addition, we are interested in detecting interesting network events from users� communication on OSN sites and studying the temporal aspects�how the various types of user feedback behave with respect to timing. We develop a novel approach which combines burst detection and text mining to detect emerging issues from online messages on a large OSN network. Through a case study, our method shows promising results in identifying a burst of activities using the OSN feedback, whereas customer care notes exhibit noticeable delays in detecting such an event which may lead to unnecessary operational expenses. }

{Abstract�Effective management of large-scale cellular data networks is critical to meet customer demands and expectations. Customer calls for technical support provides direct indication as to the issues and problems customers encounter. In this paper we study the customer tickets � free-text recordings and classifications by customer support agents � collected at a large cellular network provider, with two inter-related goals: i) to characterize and understand the major factors which lead to customers to call and seek support; and ii) to utilize such customer tickets to help identify potential network problems. For this purpose, we develop a novel statistical approach to model customer call rates which account for customer-side factors (e.g., user tenure and handset types) as well as geo-locations. We show that most calls are due to customer-side factors and can be well captured by the model. Furthermore, we also demonstrate that location-specific deviations from the model provide a good indicator of potential network-side issues. The latter is corroborated with the detailed analysis of customer tickets and other independent data sources (non-ticket customer feedback and network performance data).}

(c) ACM, 2011. This is the author's version of the work. It is posted here by permission of ACM for your personal use. Not for redistribution. The definitive version was published in ACM SIGCOMM Workshop on Measurements Up the STack (W‐MUST) , 2011-08-19.

{In this paper, we study the Location-based Reporting Tool (LRT), a smartphone application for collecting large-scale feedback from mobile customers. Using one-year data col- lected from one of the largest cellular networks in the US, we compare LRT feedback to the traditional customer feedback channel � customer care tickets. Our analysis shows that, due to the light-weight design, LRT encourages customers to report more problems from anywhere and at anytime. In addition, We find LRT users access network services more intensively than other mobile users, and thus are more likely to experience and are more sensitive to network problems. All these render LRT feedback a valuable information source for early detection of emerging network problems.}