Aaron Beach

Enhancing group recommendation by incorporating social relationship interactions

Group recommendation, which makes recommendations to a group of users instead of individuals, has become increasingly important in both the workspace and peoples social activities, such as brainstorming sessions for coworkers and social TV for family members or friends. Group recommendation is a challenging problem due to the dynamics of group memberships and diversity of group members. Previous work focused mainly on the content interests of group members and ignored the social characteristics within a group, resulting in suboptimal group recommendation performance. In this work, we propose a group recommendation method that utilizes both social and content interests of group members. We study the key characteristics of groups and propose (1) a group consensus function that captures the social, expertise, and interest dissimilarity among multiple group members; and (2) a generic framework that automatically analyzes group characteristics and constructs the corresponding group consensus function. Detailed user studies of diverse groups demonstrate the effectiveness of the proposed techniques, and the importance of incorporating both social and content interests in group recommender systems.

Fusing mobile, sensor, and social data to fully enable context-aware computing

In this paper, we identify mobile social networks as an important new direction of research in mobile computing, and show how an expanded definition of mobile social networks that includes sensor networks can enable exciting new context-aware applications, such as context-aware video screens, music jukeboxes, and mobile health applications. We offer SocialFusion as a system capable of systematically integrating such diverse mobile, social, and sensing input streams and effectuating the appropriate context-aware output action. We explain some of the major challenges that SocialFusion must overcome. We describe some preliminary results that we have obtained in implementing the SocialFusion vision.

q-Anon: Rethinking Anonymity for Social Networks

This paper proposes that social network data should be assumed public but treated private. Assuming this rather confusing requirement means that anonymity models such as k-anonymity cannot be applied to the most common form of private data release on the internet, social network APIs. An alternative anonymity model, q-Anon, is presented, which measures the probability of an attacker logically deducing previously unknown information from a social network API while assuming the data being protected may already be public information. Finally, the feasibility of such an approach is evaluated suggesting that a social network site such as Facebook could practically implement an anonymous API using q-Anon, providing its users with an anonymous option to the current application model.

Social-K: Real-time K-anonymity guarantees for social network applications

Traditional approaches to K-anonymity provide privacy guarantees over publicly released data sets with specified quasi-identifiers. However, the most common public releases of personal data are now done through social networks and their APIs, which do not fit the previous research-centric data set release model, nor do they allow for clear assumptions about quasi-identifiers. This paper proposes a new definition of K-anonymity that suggests a practical way in which social networks could provide privacy guarantees to users of their API. To support as wide a range of applications as possible, the proposed privacy guarantee assumes all social-networking data may be a quasi-identifier and does not assume that data may be generalized and still be useful. Using the Facebook social networking API, we implement an application to demonstrate that providing such guarantees in real-time is feasible for real social networking data.

SWARMS: A Sensornet Wide Area Remote Management System

Our experiences deploying a wide area wireless sensor network (WSN) in the wildfires of Idaho motivate the need for a software middleware system capable of remotely managing many sensor nodes deployed in widely disparate geographic regions. This requirement is unlike the localized focus of many traditional WSN middleware systems, which manage a group of sensor nodes deployed in a single small region, e.g. a warehouse or lab. We describe in this paper SWARMS, a wide area sensor network management system. The SWARMS architecture is designed for scalability and flexibility, while providing an infrastructure to manage in situ sensor nodes, e.g. upload code images, retrieve diagnostics, etc. To demonstrate its flexibility, we present two deployments of SWARMS, the first in a wide area weather sensor network, and the second in a local area testbed that was used by a class of graduate students. To demonstrate its scalability, we analyze the performance of SWARMS when the middleware is subject to sensor data loads of thousands of packets per second.

Touch Me Wear: Getting Physical with Social Networks

This project explores possibilities of associationbetween existingsocial network information and real-worldphysical interaction. Thesepossibilities are explored through theintegration of wearablestechnology with mobile social networktechnology. We use the term“mobile social networking” torefer to those advanced interactionswhich integrate complexreal world actions with online social networkinformation. Thisproject focuses particularly on the physical aspectsof mobilesocial networks as basic as human touch. Touch Me wE@r isableto associate social network identities with real worldphysicalcontact and in turn advertise that information online or useitreal-time within the physical world. As a proof of concept,the TouchMe wE@r application was implemented. Touch MewE@r changes the color ofone’s shirt depending on who theyhave hugged. Also the details of whowas hugged and when areadvertised on the user’s social networkprofile. One can imaginemany other interesting uses for thistechnology. The possibilitiesand implications of this technology arediscussed.

q-Anon: practical anonymity for social networks

This paper discusses why many of the common assumptions made in anonymity research cannot be applied to social network data. In particular, the concepts of ‘public’ and ‘private’ cannot be used to neatly distinguish certain types of social network data from others. It is proposed that social network data should be assumed public and treated private. An alternative anonymity model, q-Anon, is presented, which reconciles the paradox of social network data’s public/private nature. Finally, the feasibility of such an approach is evaluated suggesting that a social network site such as Facebook could practically implement an anonymous API using q-Anon. The paper concludes with a practical discussion of how q-Anon may affect different types of applications.