Antti Riikonen

Features as predictors of phone popularity

Applying Bayesian Networks, the study predicts mobile phone popularity with features.Annual predictivity elicits structural breaks which stem from changes in choice criteria.Most of discovered structural breaks relate to phone display, communication and camera.Structural break related to phone manufacturer brand is linked with market turning from hardware to software driven mode. This article analyzes dynamic changes in mobile phone popularity based on phone features. The time period, 2004-2013, selected for the study is interesting because many technical innovations took place molding the mobile phone market dramatically. The study utilizes comprehensive phone model and sales data collected in Finland, combined with temporally ordered probabilistic models to discover the time behavior of predictivity. More precisely, the Tree Augmented Naive Bayes - classification method is adapted to detect those phone characteristics that best predict the annual phone popularity measured as phone model unit sales. Linear regression and the Chow test are used to discover potential trends and structural breaks. The strength of the predictivity is measured as Kullback-Leibler Divergence. This kind of systematic longitudinal analysis highlights patterns, which are otherwise not possible to observe. The study discovered that the operating system is clearly the only feature with an increasing strength in predicting popularity over time. In contrast, sixteen features have structural breaks between 2004 and 2013. Most such breaks are related to the technical evolution of phones: their display, communication, and camera capabilities. Notably, the structural break in 2007-2008 related to the phone manufacturer brand is interpreted as the market turning from hardware to software driven mode, which contributed to Nokias failure with Symbian and Windows operating systems, and to Googles success with a hardware independent operating system Android.

The effect of an individual's age on the perceived importance and usage intensity of communications services--A Bayesian Network analysis

Multiple novel interpersonal communications services have emerged recently, but how their usage and perceived importance are related to the personal characteristics of the users is still relatively unexplored. Therefore, the aim of this study is to explore the effect of an individual’s age on the perceived importance and usage intensity of communications services based on Bayesian Networks using a survey of 3008 Finns during 2011. In the case of Short Message Service (SMS), Instant Messaging (IM), Internet forums and communities (e.g., Facebook & Twitter), and e-mail the results indicate that the perceived importance of the communications services decreases as the age increases. With phone calls and letters, however, no clear dependencies with age were identified. In the causal analysis the importance of Internet forums and communities was the only variable which can be stated to be directly caused by an individual’s age. This variable also acts as a mediator in the path from age towards perceived importance of other communication services and also towards their usage intensity. These results about the central role of Internet forums and communities can be exploited, for example, by device manufacturers when designing their products, and by service providers when designing their consumer services. The study also provides new information for mobile operators about the dependencies between mobile communications services and a documented example workflow for research community to construct a causal Bayesian Network from a combination of observational data and domain expertise.

Price and sales volume patterns of mobile handsets and technologies

This article provides empirical evidence on the price and unit sales volume patterns of mobile handsets and mobile technologies, using data on the Finnish market. The prices and sales are studied on product category, product model, and product feature levels. The results show how the dynamic of prices and sales changed after the proliferation of smartphones. Otherwise, the dynamics seem to be relatively systematic supporting the use of simple assumptions in practical estimations. The median price of handset models decreases linearly, from 89% of the introduction price at peak sales in the fifth sales month to 47% in in two years. For mobile handset features, a decreasing price pattern was also identified. After a 10% market share is reached, the decrease is on average 30 Euros per 10% market share change. The prices at feature takeoff were identified to be on average at 58% of introduction price. Price and Sales Volume Patterns of Mobile Handsets and Technologies

Mobile handset features as service enablers - Conceptual model of feature diffusion

Mobile handsets are evolving into advanced multipurpose devices with various hardware and software features enabling new services. Therefore, understanding the diffusion of mobile handset features is increasingly important. This study reviews previous research and analyzes two cases to identify the most important feature characteristics and their relationship with the feature diffusion process. In addition, a conceptual model of feature diffusion is proposed and strategies to stimulate feature diffusion are discussed.

Characterizing traffic generated with laptop computers and mobile handsets in GPRS/UMTS core networks

This paper presents a recent profile of IP data traffic in GPRS/UMTS core networks. The analysis in this paper is based on Layer 3 and 4 headers of packets which we measured in two commercial GPRS/UMTS networks at the end of 2008. We characterize traffic generated with Windows and Symbian devices which are regarded as major operating systems for laptops and mobile handsets in our datasets, respectively. Firstly, we describe the traffic characteristics by device types from the viewpoints of individual flows. After that, we introduce a notion of flow group which has a better capability to represent devicelevel usage than individual flows. In addition to presenting the differences of traffic characteristics between both device types, we also argue that the notion of flow group is beneficial for identifying operating systems of mobile devices and for alleviating uncertainties in the results of application identification according only to Layer 4 information.

A Framework for Measuring the Deployment of Internet Protocols

Internet protocols spread to potential adopters through several successive phases of implementation, commercialization, acquisition, and adoption of the protocol. This process of protocol deployment involves several stakeholders and varies depending on the deployment environment and the protocol in question. This complexity and the lack of comprehensive measurement studies call for a further conceptualization of measuring protocol diffusion along the whole deployment process. Therefore, this article develops a framework for measuring the deployment of Internet protocols, consisting of deployment steps, deployment models, deployment measures, and data sources. The measures are further linked to each other through deployment gaps and delays. In order to demonstrate the framework, it is used to assess how a set of pre-installed protocols spread in the Finnish mobile market. The framework highlights the differences between the deployment models and the importance to use both the deployment measures and gaps in the analysis of protocol success. Furthermore, the illustrative results indicate that protocol deployment is driven by applications, and show the existence of large deployment gaps between the protocol possession and usage. The results are relevant especially to researchers interested in holistically analyzing protocol deployment and protocol developers for measuring and improving the success of their protocols.