Alisdair McDiarmid

Nodobo: Mobile Phone as a Software Sensor for Social Network Research

Modern smart phones are now capable of gathering information about a users social interactions. The authors have developed and deployed Nodobo, a suite of social sensor software for Android. Our first study group is a class of senior high school students, each using a Google Nexus One mobile phone running Nodobo, which we use to capture their device usage patterns and social interactions. We provide an overview of the system architecture, describe the trial, and share some initial results.

Estimating the Strength of Ties in Communication Networks with a Small Number of Users

An accurate method of estimating the tie strengths of mobile phone users will enable the development of socially-aware mobile applications. The strength of ties between mobile phone users can be estimated for large numbers of users by using the aggregated call duration as a proxy for tie strength. This paper investigates appropriate proxies for tie strength in a network of less than one hundred people. We find that aggregated call duration is a reasonable proxy for tie strength for a small number of users, as is total call count, and that aggregated proximity duration and total proximity count are also approximate proxies for tie strength.

Achieving anonymous location-based services

Many new cell phones enable precise location measurement using GPS, and operators plan to exploit this by providing location-based services. However, the user can lose some privacy in revealing his GPS-measured location, which is an order of magnitude more precise than the network operators best estimate. We propose a protocol to allow anonymous use of location-based services, without revealing the users identity and location to any third party.

FOAF: improving detected social network accuracy

This paper presents a Bluetooth-triggered friend-of-a-friend (FOAF) presence notification application as a means to improve the accuracy of social graphs detected via mobile networks. By checking for common contacts between co-present users, it both introduces users who are not already acquainted, and improves the accuracy of the detected social graph.

Commitment-Aware Reputation System for the Digital Marketplace

The digital marketplace (DMP) connects mobile users to network service providers, who bid to provide connectivity on a per-call basis. To operate successfully, the DMP must allow users to differentiate between network providers based on past reliability. DMP contracts include an agreed commitment level, to allow providers to offer varying quality-of-service levels. We present a new reputation system which scores network providers according to how well they meet commitments. Simulation results are presented which show that the reputation function is fair, responsive, and stable

Securing the digital marketplace

Future generations of mobile systems will require novel call management strategies. The Digital Marketplace uses a market approach to manage network services, in which operators bid to provide connectivity to users. To operate successfully, the Digital Marketplace protocols must be designed securely. In this paper, we present and consider a set of security requirements, and describe countermeasures which will secure the Digital Marketplace.

Digital marketplace security requirements

Future generations of mobile systems requires novel call management strategies. The digital marketplace uses a market approach to manage network services, in which operators bid to provide connectivity to users. To operate successfully, the digital marketplace protocols must be designed securely. In this paper, we provide a threat analysis, and identify the security requirements to counter these threats.

FOAF introductions: Automatically growing and improving the accuracy of detected social graphs

Friend of a Friend (FOAF) introductions are an effective means to grow detected social graphs in mobile applications. By detecting triadic closure in the social graph they both introduce co-present users who are not already acquainted, and improve the accuracy of the social graph. We searched for the presence of triadic closure in a real mobile social network, and tested the suitability of FOAF introductions as a means for efficient network growth. We found that triadic closure does occur in mobile networks, and that half of the triads found were suitable for FOAF introductions.