Laura M. Smith

The Role of Social Media in the Discussion of Controversial Topics

In recent years, social media has revolutionized how people communicate and share information. Twitter and other blogging sites have seen an increase in political and social activism. Previous studies on the behaviors of users in politics have focused on electoral candidates and election results. Our paper investigates the role of social media in discussing and debating controversial topics. We apply sentiment analysis techniques to classify the position (for, against, neutral) expressed in a tweet about a controversial topic and use the results in our study of user behavior. Our findings suggest that Twitter is primarily used for spreading information to like-minded people rather than debating issues. Users are quicker to rebroadcast information than to address a communication by another user. Individuals typically take a position on an issue prior to posting about it and are not likely to change their tweeting opinion.

Improving density estimation by incorporating spatial information

Given discrete event data, we wish to produce a probability density that can model the relative probability of events occurring in a spatial region. Common methods of density estimation, such as Kernel Density Estimation, do not incorporate geographical information. Using these methods could result in nonnegligible portions of the support of the density in unrealistic geographic locations. For example, crime density estimation models that do not take geographic information into account may predict events in unlikely places such as oceans, mountains, and so forth. We propose a set of Maximum Penalized Likelihood Estimation methods based on Total Variation and Sobolev norm regularizers in conjunction with a priori high resolution spatial data to obtain more geographically accurate density estimates. We apply this method to a residential burglary data set of the San Fernando Valley using geographic features obtained from satellite images of the region and housing density information.

Spectral clustering with epidemic diffusion.

Spectral clustering is widely used to partition graphs into distinct modules or communities. Existing methods for spectral clustering use the eigenvalues and eigenvectors of the graph Laplacian, an operator that is closely associated with random walks on graphs. We propose a spectral partitioning method that exploits the properties of epidemic diffusion. An epidemic is a dynamic process that, unlike the random walk, simultaneously transitions to all the neighbors of a given node. We show that the replicator, an operator describing epidemic diffusion, is equivalent to the symmetric normalized Laplacian of a reweighted graph with edges reweighted by the eigenvector centralities of their incident nodes. Thus, more weight is given to edges connecting more central nodes. We describe a method that partitions the nodes based on the componentwise ratio of the replicators second eigenvector to the first and compare its performance to traditional spectral clustering techniques on synthetic graphs with known community structure. We demonstrate that the replicator gives preference to dense, clique-like structures, enabling it to more effectively discover communities that may be obscured by dense intercommunity linking.

Partitioning Networks with Node Attributes by Compressing Information Flow

Real-world networks are often organized as modules or communities of similar nodes that serve as functional units. These networks are also rich in content, with nodes having distinguished features or attributes. In order to discover a network’s modular structure, it is necessary to take into account not only its links but also node attributes. We describe an information-theoretic method that identifies modules by compressing descriptions of information flow on a network. Our formulation introduces node content into the description of information flow, which we then minimize to discover groups of nodes with similar attributes that also tend to trap the flow of information. The method is conceptually simple and does not require ad-hoc parameters to specify the number of modules or to control the relative contribution of links and node attributes to network structure. We apply the proposed method to partition real-world networks with known community structure. We demonstrate that adding node attributes helps recover the underlying community structure in content-rich networks more effectively than using links alone. In addition, we show that our method is faster and more accurate than alternative state-of-the-art algorithms.

Using Stochastic Models to Predict User Response in Social Media

User response to contributed content in online social media depends on many factors. These include how the site lays out new content, how frequently the user visits the site, how many friends the user follows, how active these friends are, as well as how interesting or useful the content is to the user. We present a stochastic modeling framework that relates a users behavior to details of the sites user interface and user activity and describe a procedure for estimating model parameters from available data. We apply the model to study discussions of controversial topics on Twitter, specifically, to predict how followers of an advocate for a topic respond to the advocates posts. We show that a model of user behavior that explicitly accounts for a user discovering the advocates post by scanning through a list of newer posts, better predicts response than models that do not.

Modelling policing strategies for departments with limited resources

Crime prevention is a major goal of law-enforcement agencies. Often, these agencies have limited resources and officers available for patrolling and responding to calls. However, patrolling and police visibility can influence individuals to not perform criminal acts. Therefore, it is necessary for the police to optimize their patrolling strategies to deter the most crime. Previous studies have created agent-based models to simulate criminal and police agents interacting in a city, indicating a “cops on the dots” strategy as a viable method to mitigate large amounts of crime. Unfortunately, police departments cannot allocate all of the patrolling officers to seek out these hotspots, particularly since they are not immediately known. In large cities, it is often necessary to keep a few officers in different areas of the city, frequently divided up into beats. Officers need to respond to calls, possibly not of a criminal nature. Therefore, we modify models for policing to account for these factors. Through testing the policing strategies for various hotspot types and number of police agents, we found that the methods that performed the best varied greatly according to these factors.

Multi-stage Vector-Borne Zoonoses Models: A Global Analysis

A class of models that describes the interactions between multiple host species and an arthropod vector is formulated and its dynamics investigated. A host-vector disease model where the host’s infection is structured into n stages is formulated and a complete global dynamics analysis is provided. The basic reproduction number acts as a sharp threshold, that is, the disease-free equilibrium is globally asymptotically stable (GAS) whenever

Stochastic Models Predict User Behavior in Social Media

{\mathcal {R}}_0^2\le 1