Gianluca Lax

Discovering links among social networks

Distinct social networks are interconnected via bridge users, who play thus a key role when crossing information is investigated in the context of Social Internetworking analysis. Unfortunately, not always users make their role of bridge explicit by specifying the so-called me edge (i.e., the edge connecting the accounts of the same user in two distinct social networks), missing thus a potentially very useful information. As a consequence, discovering missing me edges is an important problem to face in this context yet not so far investigated. In this paper, we propose a common-neighbors approach to detecting missing me edges, which returns good results in real life settings. Indeed, an experimental campaign shows both that the state-of-the-art common-neighbors approaches cannot be effectively applied to our problem and, conversely, that our approach returns precise and complete results.

Moving from social networks to social internetworking scenarios: The crawling perspective

In new generation social networks, we expect that the paradigm of Social Internetworking Systems (SISs) will become progressively more important. Indeed, the possibility of interconnecting users and resources of different social networks enables a lot of strategic applications whose main strength is the integration of different communities that nevertheless preserves their diversity and autonomy. In this new scenario, the role of Social Network Analysis is crucial in studying the evolution of structures, individuals, interactions, and so on, and in extracting powerful knowledge from them. But the preliminary step to do is designing a good way to crawl the underlying graph. Although this aspect has been deeply investigated in the field of social networks, it is an open issue when moving towards SISs. Indeed, we cannot expect that a crawling strategy, specifically designed for social networks, is still valid in a Social Internetworking Scenario, due to its specific topological features. In this paper, we confirm the above claim, giving a strong motivation for our second contribution, which is the definition of a new crawling strategy. This strategy, specifically conceived for SISs, is shown to fully overcome the drawbacks of the state-of-the-art crawling strategies.

Discovering missing me edges across social networks

Distinct social networks are interconnected via membership overlap, which plays a key role when crossing information is investigated in the context of multiple-social-network analysis. Unfortunately, users do not always make their membership to two distinct social networks explicit, by specifying the so-called me edge (practically, corresponding to a link between the two accounts), thus missing a potentially very useful information. As a consequence, discovering missing me edges is an important problem to address in this context with potential powerful applications. In this paper, we propose a common-neighbor approach to detecting missing me edges, which returns good results in real-life settings. Indeed, an experimental campaign shows both that the state-of-the-art common-neighbor approaches cannot be effectively applied to our problem and, conversely, that our approach returns precise and complete results.

Comparing Twitter and Facebook user behavior

We compare the behavior of users who belong to both Twitter and Facebook.We adopt a truly multi-OSN perspective by basing our analysis on membership overlap.We study overlapping friendship, user activity, degree, and privacy awareness. Understanding online-social-network (OSN) user behavior is an important challenge in the field of social network analysis, as OSNs play a significant role in peoples daily lives. So far, many studies considering only one OSN or, at most, comparing results obtained for a single OSN, have been provided. Nowadays, users typically join more OSNs and this is an important aspect that should be taken into account for user behavior analysis. In this paper, we give an important contribution in this direction, by analyzing the behavior of users belonging to both Facebook and Twitter. This way, the analysis is well-founded because it is conducted on a common set of users and, further, a number of specific analyses become possible (as common friendship). Our study is carried out on data extracted from the web, and allows us to find important specificities of these users about their privacy setting, the choice of friends and the activity they do, which are generally consistent with the recent findings in this field.

CellTrust: a reputation model for C2C commerce

The improvement of wireless technologies and the increasing spread of mobile phones open new possibilities to perform mobile Customer-to-Customer commercial activities. In this new scenario, where users cannot rely on stable connections, it assumes a great relevance how to trust the counterpart in a transaction and how to avoid that a disconnection, possible in wireless connections, can encourage users to cheat. To tackle these issues we propose a feedback-based reputation mechanism able to detect malicious users better than other state-of-the-art techniques, as shown by the large number of experiments run to measure the accuracy of the compared methods in the most common situations.

Implementing disposable credit card numbers by mobile phones

Disposable credit card numbers are a recent approach to tackling the severe problem of credit card fraud, nowadays constantly growing, especially in the context of e-commerce payments. Whenever we cannot rely on a secure communication channel between cardholder and issuer, a possibility is to generate new numbers on the basis of some common scheme, starting from a shared secret information. However, in order to make the approach meaningful from a practical point of view, the solution should guarantee backward compatibility with the current system, absence of new investments in dedicated hardware, wide-spectrum usability, and adequate security level. In this paper, we propose a solution based on the use of standard mobile phones, fully meeting the above desiderata. Importantly, our solution does not require any cryptographic support and, as a consequence, the use of PADs or smart phones, opening then its usability to a wider potential market.

Crawling Social Internetworking Systems

In new generation social networks, we expect that the paradigm of Social Internetworking Systems (SISs, for short) will be more and more important. In this new scenario, the role of Social Network Analysis is of course still crucial but the preliminary step to do is designing a good way to crawl the underlying graph. While this aspect has been deeply investigated in the field of social networks, it is an open issue when moving towards SISs. Indeed, we cannot expect that a crawling strategy which is good for social networks, is still valid in a Social Internetworking Scenario, due to its specific topological features. In this paper, we first confirm the above claim and, then, define a new crawling strategy specifically conceived for SISs. Finally, we show that it fully overcomes the drawbacks of the state-of-the-art crawling strategies.

Enhancing histograms by tree-like bucket indices

Histograms are used to summarize the contents of relations into a number of buckets for the estimation of query result sizes. Several techniques have been proposed in the past for determining bucket boundaries which provide accurate estimations. However, while search strategies for optimal bucket boundaries are rather sophisticated, no much attention has been paid for estimating queries inside buckets and all of the above techniques adopt naive methods for such an estimation. This paper focuses on the problem of improving the estimation inside a bucket once its boundaries have been fixed. The proposed technique is based on the addition, to each bucket, of a memory-word additional information (organized into a tree-like index), storing approximate cumulative frequencies in a hierarchical fashion. Both theoretical analysis and experimental results show that the proposed approach improves the accuracy of the estimation inside buckets, w.r.t. both classical approaches (like continuous value assumption and uniform spread assumption) and a number of alternative ways to organize the additional information. The index is later added to state-of-the-art histograms obtaining the non-obvious result that despite the spatial overhead which reduces the number of allowed buckets once the storage space has been fixed, the original methods are strongly improved in terms of accuracy.

Driving Global Team Formation in Social Networks to Obtain Diversity

In this paper, we present a preliminary idea for a crowdsourcing application aimed at driving the process of global team formation to obtain diversity in the team. Indeed, it is well known that diversity is one of the key factors of collective intelligence in crowdsourcing. The idea is based on the identification of suitable nodes in social networks, which can profitably play the role of generators of diversity in the team formation process. This paper presents a first step towards the concrete definition of the above application consisting in the identification of an effective measure that can be used to select the most promising nodes w.r.t. the above feature.

A model to support design and development of multiple-social-network applications

Online social networks have become so pervasive in peoples lives that they can play a crucial role in design and development processes of applications. At moment, a gap exists w.r.t. standard networking programming to support social-network-based programming in large, according to software engineering principles of genericity and polymorphism. This drawback is made evident when applications should be built on top of multiple social networks and the user-centered vision should be kept. Indeed, heterogeneity of social networks does not allow us to produce software with suitable abstraction. In this paper, we cover the above gap by defining and implementing a model aimed at generalizing concepts, actions and relationships of existing social networks. The effectiveness of our approach is shown by two case studies.