arXiv subjects

Don Towsley

Publications and source records attributed to Don Towsley.

At least 199 records · Page 11Linked to original sources

arXiv2014

Who is Dating Whom: Characterizing User Behaviors of a Large Online Dating Site

Online dating sites have become popular platforms for people to look for potential romantic partners. It is important to understand users' dating preferences in order to make better recommendations on potential dates. The message sending and replying actions of a user are strong indicators for what he/she is looking for in a potential date and reflect the user's actual dating preferences. We study how users' online dating behaviors correlate with various user attributes using a large real-world dateset from a major online dating site in China. Many of our results on user messaging behavior align with notions in social and evolutionary psychology: males tend to look for younger females while females put more emphasis on the socioeconomic status (e.g., income, education level) of a potential date. In addition, we observe that the geographic distance between two users and the photo count of users play an important role in their dating behaviors. Our results show that it is important to differentiate between users' true preferences and random selection. Some user behaviors in choosing attributes in a potential date may largely be a result of random selection. We also find that both males and females are more likely to reply to users whose attributes come closest to the stated preferences of the receivers, and there is significant discrepancy between a user's stated dating preference and his/her actual online dating behavior. These results can provide valuable guidelines to the design of a recommendation engine for potential dates.

cs.SI↗

arXiv2013

Multi-Source Multi-Path HTTP (mHTTP): A Proposal

Today, most devices have multiple network interfaces. Coupled with wide-spread replication of popular content at multiple locations, this provides substantial path diversity in the Internet. We propose Multi-source Multipath HTTP, mHTTP, which takes advantage of all existing types of path diversity in the Internet. mHTTP needs only client-side but not server-side or network modifications as it is a receiver-oriented mechanism. Moreover, the modifications are restricted to the socket interface. Thus, no changes are needed to the applications or to the kernel. As mHTTP relies on HTTP range requests, it is specific to HTTP which accounts for more than 60% of the Internet traffic. We implement mHTTP and study its performance by conducting measurements over a testbed and in the wild. Our results show that mHTTP indeed takes advantage of all types of path diversity in the Internet, and that it is a viable alternative to Multipath TCP for HTTP traffic. mHTTP decreases download times for large objects up to 50%, whereas it does no harm to small object downloads.

cs.NI↗

arXiv2013

Social Sensor Placement in Large Scale Networks: A Graph Sampling Perspective

Sensor placement for the purpose of detecting/tracking news outbreak and preventing rumor spreading is a challenging problem in a large scale online social network (OSN). This problem is a kind of subset selection problem: choosing a small set of items from a large population so to maximize some prespecified set function. However, it is known to be NP-complete. Existing heuristics are very costly especially for modern OSNs which usually contain hundreds of millions of users. This paper aims to design methods to find \emph{good solutions} that can well trade off efficiency and accuracy. We first show that it is possible to obtain a high quality solution with a probabilistic guarantee from a "{\em candidate set}" of the underlying social network. By exploring this candidate set, one can increase the efficiency of placing social sensors. We also present how this candidate set can be obtained using "{\em graph sampling}", which has an advantage over previous methods of not requiring the prior knowledge of the complete network topology. Experiments carried out on two real datasets demonstrate not only the accuracy and efficiency of our approach, but also effectiveness in detecting and predicting news outbreak.

cs.SI↗

arXiv2013

Practical Characterization of Large Networks Using Neighborhood Information

Characterizing large online social networks (OSNs) through node querying is a challenging task. OSNs often impose severe constraints on the query rate, hence limiting the sample size to a small fraction of the total network. Various ad-hoc subgraph sampling methods have been proposed, but many of them give biased estimates and no theoretical basis on the accuracy. In this work, we focus on developing sampling methods for OSNs where querying a node also reveals partial structural information about its neighbors. Our methods are optimized for NoSQL graph databases (if the database can be accessed directly), or utilize Web API available on most major OSNs for graph sampling. We show that our sampling method has provable convergence guarantees on being an unbiased estimator, and it is more accurate than current state-of-the-art methods. We characterize metrics such as node label density estimation and edge label density estimation, two of the most fundamental network characteristics from which other network characteristics can be derived. We evaluate our methods on-the-fly over several live networks using their native APIs. Our simulation studies over a variety of offline datasets show that by including neighborhood information, our method drastically (4-fold) reduces the number of samples required to achieve the same estimation accuracy of state-of-the-art methods.

cs.SI↗

arXiv2013

Sampling Content Distributed Over Graphs

Despite recent effort to estimate topology characteristics of large graphs (i.e., online social networks and peer-to-peer networks), little attention has been given to develop a formal methodology to characterize the vast amount of content distributed over these networks. Due to the large scale nature of these networks, exhaustive enumeration of this content is computationally prohibitive. In this paper, we show how one can obtain content properties by sampling only a small fraction of vertices. We first show that when sampling is naively applied, this can produce a huge bias in content statistics (i.e., average number of content duplications). To remove this bias, one may use maximum likelihood estimation to estimate content characteristics. However our experimental results show that one needs to sample most vertices in the graph to obtain accurate statistics using such a method. To address this challenge, we propose two efficient estimators: special copy estimator (SCE) and weighted copy estimator (WCE) to measure content characteristics using available information in sampled contents. SCE uses the special content copy indicator to compute the estimate, while WCE derives the estimate based on meta-information in sampled vertices. We perform experiments to show WCE and SCE are cost effective and also ``{\em asymptotically unbiased}''. Our methodology provides a new tool for researchers to efficiently query content distributed in large scale networks.

cs.SI↗

arXiv2013

Enabling Information Centric Networks through Opportunistic Search, Routing and Caching

Content dissemination networks are pervasive in todays Internet. Examples of content dissemination networks include peer-to-peer networks (P2P), content distribution networks (CDN) and information centric networks (ICN). In this paper, we propose a new system design for information centric networks which leverages opportunistic searching, routing and caching. Our system design is based on an hierarchical tiered structure. Random walks are used to find content inside each tier, and gateways across tiers are used to direct requests towards servers placed in the top tier, which are accessed in case content replicas are not found in lower tiers. Then, we propose a model to analyze the system in consideration. The model yields metrics such as mean time to find a content and the load experienced by custodians as a function of the network topology. Using the model, we identify trade-offs between these two metrics, and numerically show how to find the optimal time to live of the random walks.

cs.NI↗

arXiv2013

On the Scalability of Reliable Data Transfer in High Speed Networks

This paper considers reliable data transfer in a high-speed network (HSN) in which the per-connection capacity is very large. We focus on sliding window protocols employing selective repeat for reliable data transfer and study two reliability mechanisms based on ARQ and FEC. The question we ask is which mechanism is more suitable for an HSN in which the scalability of reliable data transfer in terms of receiver's buffer requirement and achievable delay and throughput is a concern. To efficiently utilize the large bandwidth available to a connection in an HSN, sliding window protocols require a large transmission window. In this regime of large transmission windows, we show that while both mechanisms achieve the same asymptotic throughput in the presence of packet losses, their delay and buffer requirements are different. Specifically, an FEC-based mechanism has delay and receiver's buffer requirement that are asymptotically smaller than that of an ARQ-based selective repeat mechanism by a factor of log W, where W is the window size of the selective repeat mechanism. This result is then used to investigate the implications of each reliability mechanism on protocol design in an HSN in terms of throughput, delay, buffer requirement, and control overhead.

cs.NI↗

arXiv2013

Limits of Reliable Communication with Low Probability of Detection on AWGN Channels

We present a square root limit on the amount of information transmitted reliably and with low probability of detection (LPD) over additive white Gaussian noise (AWGN) channels. Specifically, if the transmitter has AWGN channels to an intended receiver and a warden, both with non-zero noise power, we prove that $o(\sqrt{n})$ bits can be sent from the transmitter to the receiver in $n$ channel uses while lower-bounding $α+β\geq1-ε$ for any $ε>0$, where $α$ and $β$ respectively denote the warden's probabilities of a false alarm when the sender is not transmitting and a missed detection when the sender is transmitting. Moreover, in most practical scenarios, a lower bound on the noise power on the channel between the transmitter and the warden is known and $O(\sqrt{n})$ bits can be sent in $n$ LPD channel uses. Conversely, attempting to transmit more than $O(\sqrt{n})$ bits either results in detection by the warden with probability one or a non-zero probability of decoding error at the receiver as $n\rightarrow\infty$.

cs.IT↗

arXiv2013

Computing Traversal Times on Dynamic Markovian Paths

In source routing, a complete path is chosen for a packet to travel from source to destination. While computing the time to traverse such a path may be straightforward in a fixed, static graph, doing so becomes much more challenging in dynamic graphs, in which the state of an edge in one time slot (i.e., its presence or absence) is random, and may depend on its state in the previous time step. The traversal time is due to both time spent waiting for edges to appear and time spent crossing them once they become available. We compute the expected traversal time (ETT) for a dynamic path in a number of special cases of stochastic edge dynamics models, and for three edge failure models, culminating in a surprisingly challenging yet realistic setting in which the initial configuration of edge states for the entire path is known. We show that the ETT for this "initial configuration" setting can be computed in quadratic time, by an algorithm based on probability generating functions. We also give several linear-time upper and lower bounds on the ETT.

cs.NI↗

arXiv2013

Characterizing Branching Processes from Sampled Data

Branching processes model the evolution of populations of agents that randomly generate offsprings. These processes, more patently Galton-Watson processes, are widely used to model biological, social, cognitive, and technological phenomena, such as the diffusion of ideas, knowledge, chain letters, viruses, and the evolution of humans through their Y-chromosome DNA or mitochondrial RNA. A practical challenge of modeling real phenomena using a Galton-Watson process is the offspring distribution, which must be measured from the population. In most cases, however, directly measuring the offspring distribution is unrealistic due to lack of resources or the death of agents. So far, researchers have relied on informed guesses to guide their choice of offspring distribution. In this work we propose two methods to estimate the offspring distribution from real sampled data. Using a small sampled fraction of the agents and instrumented with the identity of the ancestors of the sampled agents, we show that accurate offspring distribution estimates can be obtained by sampling as little as 14% of the population.

stat.AP↗

arXiv2013

Asymptotic Optimality of Equal Power Allocation for Linear Estimation of WSS Random Processes

This letter establishes the asymptotic optimality of equal power allocation for measurements of a continuous wide-sense stationary (WSS) random process with a square-integrable autocorrelation function when linear estimation is used on equally-spaced measurements with periodicity meeting the Nyquist criterion and with the variance of the noise on any sample inversely proportional to the power expended by the user to obtain that measurement.

cs.IT↗

arXiv2012

Online Myopic Network Covering

Efficient marketing or awareness-raising campaigns seek to recruit $n$ influential individuals -- where $n$ is the campaign budget -- that are able to cover a large target audience through their social connections. So far most of the related literature on maximizing this network cover assumes that the social network topology is known. Even in such a case the optimal solution is NP-hard. In practice, however, the network topology is generally unknown and needs to be discovered on-the-fly. In this work we consider an unknown topology where recruited individuals disclose their social connections (a feature known as {\em one-hop lookahead}). The goal of this work is to provide an efficient greedy online algorithm that recruits individuals as to maximize the size of target audience covered by the campaign. We propose a new greedy online algorithm, Maximum Expected $d$-Excess Degree (MEED), and provide, to the best of our knowledge, the first detailed theoretical analysis of the cover size of a variety of well known network sampling algorithms on finite networks. Our proposed algorithm greedily maximizes the expected size of the cover. For a class of random power law networks we show that MEED simplifies into a straightforward procedure, which we denote MOD (Maximum Observed Degree). We substantiate our analytical results with extensive simulations and show that MOD significantly outperforms all analyzed myopic algorithms. We note that performance may be further improved if the node degree distribution is known or can be estimated online during the campaign.

cs.SI↗

arXiv2012

Characterizing Continuous Time Random Walks on Time Varying Graphs

In this paper we study the behavior of a continuous time random walk (CTRW) on a stationary and ergodic time varying dynamic graph. We establish conditions under which the CTRW is a stationary and ergodic process. In general, the stationary distribution of the walker depends on the walker rate and is difficult to characterize. However, we characterize the stationary distribution in the following cases: i) the walker rate is significantly larger or smaller than the rate in which the graph changes (time-scale separation), ii) the walker rate is proportional to the degree of the node that it resides on (coupled dynamics), and iii) the degrees of node belonging to the same connected component are identical (structural constraints). We provide examples that illustrate our theoretical findings.

cs.SI↗

arXiv2012

On Set Size Distribution Estimation and the Characterization of Large Networks via Sampling

In this work we study the set size distribution estimation problem, where elements are randomly sampled from a collection of non-overlapping sets and we seek to recover the original set size distribution from the samples. This problem has applications to capacity planning, network theory, among other areas. Examples of real-world applications include characterizing in-degree distributions in large graphs and uncovering TCP/IP flow size distributions on the Internet. We demonstrate that it is hard to estimate the original set size distribution. The recoverability of original set size distributions presents a sharp threshold with respect to the fraction of elements that remain in the sets. If this fraction remains below a threshold, typically half of the elements in power-law and heavier-than-exponential-tailed distributions, then the original set size distribution is unrecoverable. We also discuss practical implications of our findings.

math.ST↗

arXiv2012

Multiple Random Walks to Uncover Short Paths in Power Law Networks

Consider the following routing problem in the context of a large scale network $G$, with particular interest paid to power law networks, although our results do not assume a particular degree distribution. A small number of nodes want to exchange messages and are looking for short paths on $G$. These nodes do not have access to the topology of $G$ but are allowed to crawl the network within a limited budget. Only crawlers whose sample paths cross are allowed to exchange topological information. In this work we study the use of random walks (RWs) to crawl $G$. We show that the ability of RWs to find short paths bears no relation to the paths that they take. Instead, it relies on two properties of RWs on power law networks: 1) RW's ability observe a sizable fraction of the network edges; and 2) an almost certainty that two distinct RW sample paths cross after a small percentage of the nodes have been visited. We show promising simulation results on several real world networks.

cs.SI↗

arXiv2012

Quick Detection of Nodes with Large Degrees

Our goal is to quickly find top $k$ lists of nodes with the largest degrees in large complex networks. If the adjacency list of the network is known (not often the case in complex networks), a deterministic algorithm to find a node with the largest degree requires an average complexity of O(n), where $n$ is the number of nodes in the network. Even this modest complexity can be very high for large complex networks. We propose to use the random walk based method. We show theoretically and by numerical experiments that for large networks the random walk method finds good quality top lists of nodes with high probability and with computational savings of orders of magnitude. We also propose stopping criteria for the random walk method which requires very little knowledge about the structure of the network.

cs.DS↗

arXiv2011

Network Characteristics of Video Streaming Traffic

Video streaming represents a large fraction of Internet traffic. Surprisingly, little is known about the network characteristics of this traffic. In this paper, we study the network characteristics of the two most popular video streaming services, Netflix and YouTube. We show that the streaming strategies vary with the type of the application (Web browser or native mobile application), and the type of container (Silverlight, Flash, or HTML5) used for video streaming. In particular, we identify three different streaming strategies that produce traffic patterns from non-ack clocked ON-OFF cycles to bulk TCP transfer. We then present an analytical model to study the potential impact of these streaming strategies on the aggregate traffic and make recommendations accordingly.

cs.NI↗

arXiv2011

Broadcast Analysis for Large Cooperative Wireless Networks

The capability of nodes to broadcast their message to the entire wireless network when nodes employ cooperation is considered. We employ an asymptotic analysis using an extended random network setting and show that the broadcast performance strongly depends on the path loss exponent of the medium. In particular, as the size of the random network grows, the probability of broadcast in a one-dimensional network goes to zero for path loss exponents larger than one, and goes to a nonzero value for path loss exponents less than one. In two-dimensional networks, the same behavior is observed for path loss exponents above and below two, respectively.

cs.IT↗