Frank Nagle

Can Friends Be Trusted? Exploring Privacy in Online Social Networks

In this paper, we present a case study describing the privacy and trust that exist within a small population of online social network users. We begin by formally characterizing different graphs in social network sites like Facebook. We then determine how often people are willing to divulge personal details to an unknown online user, an adversary. While most users in our sample did not share sensitive information when asked by an adversary, we found that more users were willing to divulge personal details to an adversary if there is a mutual friend connected to the adversary and the user. We then summarize the results and observations associated with this Facebook case study.

Exploring re-identification risks in public domains

While re-identification of sensitive data has been studied extensively, with the emergence of online social networks and the popularity of digital communications, the ability to use public data for re-identification has increased. This work begins by presenting two different cases studies for sensitive data re-identification. We conclude that targeted re-identification using traditional variables is not only possible, but fairly straightforward given the large amount of public data available. However, our first case study also indicates that large-scale re-identification is less likely. We then consider methods for agencies such as the Census Bureau to identify variables that cause individuals to be vulnerable without testing all combinations of variables. We show the effectiveness of different strategies on a Census Bureau data set and on a synthetic data set.

Privacy Breach Analysis in Social Networks

This chapter addresses various aspects of analyzing privacy breaches in social networks. We first review literature that defines three types of privacy breaches in social networks: interactive, active, and passive. We then survey the various network anonymization schemes that have been constructed to address these privacy breaches. After exploring these breaches and anonymization schemes, we evaluate a measure for determining the level of anonymity inherent in a network graph based on its topological structure. Finally, we close by emphasizing the difficulty of anonymizing social network data while maintaining usability for research purposes and offering areas for future work.

EWNI: efficient anonymization of vulnerable individuals in social networks

Social networks, patient networks, and email networks are all examples of graphs that can be studied to learn about information diffusion, community structure and different system processes; however, they are also all examples of graphs containing potentially sensitive information. While several anonymization techniques have been proposed for social network data publishing, they all apply the anonymization procedure on the entire graph. Instead, we propose a local anonymization algorithm that focuses on obscuring structurally important nodes that are not well anonymized, thereby reducing the cost of the overall anonymization procedure. Based on our experiments, we observe that we reduce the cost of anonymization by an order of magnitude while maintaining, and even improving, the accuracy of different graph centrality measures, e.g. degree and betweenness, when compared to another well known data publishing approach.

Open Source Software and Firm Productivity

As open source software (OSS) is increasingly used as a key input by firms, understanding its impact on productivity becomes critical. This study measures the firm-level productivity impact of non-pecuniary (free) OSS and finds a positive and significant value-added return for firms in IT-producing industries. For firms in non-IT-producing industries, the results show a positive effect that is only marginally significant and can take up to three years to accrue, indicating that an ecosystem of complements is necessary to obtain the full impact. Dynamic panel analysis and a variety of robustness checks are used to add support for a causal interpretation. For IT-producing firms, a 1% increase in the use of non-pecuniary OSS leads to an increase in value-added productivity in the range of 0.003% to 0.022%. This effect is smaller for larger firms and the results indicate that prior research underestimates the amount of IT firm’s use.

Jack of All Trades and Master of Knowledge:The Role of Generalists in Novel Knowledge Integration

Organization-level knowledge diversification facilitates exploration – integration of external new knowledge –, yet knowledge accumulation poses a challenge because there is a trade-off between individual-level breadth and depth of knowledge. This leads to a need to coordinate larger teams in order to gather enough diverse expertise and capitalize on its benefits, a complex and costly process. As an alternative, we consider and show evidence of the role of individual-level diversification as a mechanism through which skilled researchers engage in successful exploration by utilizing the benefits of their breadth of knowledge and by mitigating the perceived disadvantages of their shallower depth of knowledge through diverse collaboration networks. Our results suggest that organizations seeking to innovate at the frontier should consider the benefits of hiring diverse researchers.

Online Word of Mouth and Product Review Disagreement

Studies of online word of mouth have frequently posited ― but never systematically conceptualized and explored ― that the level of disagreement between existing product reviews can impact the volume and the valence of future reviews. In this study we develop a theoretical framework of disagreement in online WOM and test our predictions in a dataset of nearly 300,000 online reviews for 425 movies over three years. This framework highlights that rather than thinking of disagreement as dispersion of opinions around a mean, high levels of disagreement can be better conceptualized as opposing opinion poles. Such a conceptualization has important implications for how disagreement can be measured and how results can be interpreted. We theoretically develop, validate, and apply a novel statistical measure of disagreement that can be used alongside existing alternative approaches such as standard deviation. We find that only high levels of disagreement ― with opposing opinion poles ― influence future reviews while simple dispersion does not. We show that high levels of disagreement among previously posted reviews lead to more future product reviews, a relationship that is moderated by informational content such that higher informational content amplifies the effect. Further, we show that increased disagreement leads to future reviews of lower valence. Our findings highlight that an important role for research on big data in information systems is to examine how existing measurement approaches and interpretations can be improved by fully leveraging the richness that digital trace data offers.