Kleinberg's Small-world Model Research Articles

Low-Distortion Inference of Latent Similarities from a Multiplex Social Network

Much of social network analysis is---implicitly or explicitly---predicated on the assumption that individuals tend to be more similar to their friends than to strangers. Thus, an observed social network provides a noisy signal about the latent underlying “social space''---the way in which individuals are similar or dissimilar. Many research questions frequently addressed via social network analysis are in reality questions about this social space, raising the question of inverting the process: Given a social network, how accurately can we reconstruct the social structure of similarities and dissimilarities? We begin to address this problem formally. Observed social networks are usually multiplex, in the sense that they reflect (dis)similarities in several different “categories,” such as geographical proximity, kinship, or similarity of professions/hobbies. We assume that each such category is characterized by a latent metric capturing (dis)similarities in this category. Each category gives rise to a separate social network: a random graph parameterized by this metric. For a concrete model, we consider Kleinberg's small world model and some variations thereof. The observed social network is the unlabeled union of these graphs; i.e., the presence or absence of edges can be observed, but not their origins. Our main result is an efficient algorithm which reconstructs each metric with provably low distortion.

Study on Dissemination Patterns in Location-Aware Gossiping Networks

We study the properties of information dissemination over location-aware gossiping networks leveraging location-based real-time communication applications. Gossiping is a promising method for quickly disseminating messages in a large-scale system, but in its application to information dissemination for location-aware applications, it is important to consider the network topology and patterns of spatial dissemination over the network in order to achieve effective delivery of messages to potentially interested users. To this end, we propose a continuous-space network model extended from Kleinberg's small-world model applicable to actual location-based applications. Analytical and simulation-based study shows that the proposed network achieves high dissemination efficiency resulting from geographically neutral dissemination patterns as well as selective dissemination to proximate users. We have designed a highly scalable location management method capable of promptly updating the network topology in response to node movement and have implemented a distributed simulator to perform dynamic target pursuit experiments as one example of applications that are the most sensitive to message forwarding delay. The experimental results show that the proposed network surpasses other types of networks in pursuit efficiency and achieves the desirable dissemination patterns.

Mechanism of a Resource Location Approach in Virtual Computing Systems

We consider the problem of how to effectively locate resources in virtual computing systems, such as grid and cloud computing environments. In this paper, on the base of Kleinberg Small World model, aimed power-law characteristics, an efficient decentralized construction approach PLSWCP(Power-law oriented Small World Construction Protocol) is presented, which uses fairly small partial view instead of global knowledge of network, getting intelligent choices in complicated environments. Theoretical analysis and simulations show that PLSWCP is scalable, self-adaptable, and load-balanced, improving the efficiency of resource location.

Mechanism of a Resource Location Approach in Power-Law Networks

How to effectively locate resources is a very important factor affecting the performance of distributed system in wide area network environments. Some resource location methods have been already proposed, which utilize Small World phenomena, but have not show how to construct a Small World exactly. In this paper, on the base of Kleinberg Small World model, aimed power-law characteristics, an efficient decentralized construction approach PLSWCP(Power-law oriented Small World Construction Protocol) is proposed, which uses fairly small partial view instead of global knowledge of network. Theoretical analysis and simulations show that PLSWCP is scalable, self-adaptable, and load-balanced, improving the efficiency of resource location.