On Latent Position Inference from Doubly Stochastic Messaging Activities

  title={On Latent Position Inference from Doubly Stochastic Messaging Activities},
  author={Nam H. Lee and Jordan Yoder and Minh Tang and Carey E. Priebe},
  journal={Multiscale Model. Simul.},
We model messaging activities as a hierarchical doubly stochastic point process with three main levels, and develop an iterative algorithm for inferring actors' relative latent positions from a stream of messaging activity data. Each of the message-exchanging actors is modeled as a process in a latent space. The actors' latent positions are assumed to be influenced by the distribution of a much larger population over the latent space. Each actor's movement in the latent space is modeled as… 

Point-process models of social network interactions: Parameter estimation and missing data recovery

A relaxed maximum likelihood method for filling in missing data in records of communications in social networks and its optimisation algorithm adapts a recent curvilinear search method to handle inequality constraints and a non-vanishing derivative.

Spectral embedding for dynamic networks with stability guarantees

This work argues that two types of stability in the spatio-temporal positioning of nodes are desirable: to assign the same position to nodes behaving similarly at a given time (cross-sectional stability) and a constant position, up to noise, to a single node behaving similarly across different times (longitudinal stability).

Mathematical Models and Methods for Behavior in Social Networks: Urban Crime, Self-Exciting Interactions, and Information Spread

This work considers the problem of filling in missing data in records of communication, formulating it as a variational problem and developing a numerical method to solve it, and adopts a point-process model to reflect observed temporal bursts in communication.

Community Detection using Locality Statistics

A novel framework for detecting active communities that consist of the most active vertices is proposed, which utilizes a parallelizable trimming algorithm based on a locality statistic to filter out inactive vertices, and then clusters the remaining active Vertices via spectral decomposition of their vertices.

Locality Statistics for Anomaly Detection in Time Series of Graphs

This paper formulates change-point detection as a hypothesis testing problem in terms of a generative latent position model, focusing on the special case of the Stochastic Block Model time series.

An integrative framework for sensor-based measurement of teamwork in healthcare

Initial evidence exists to support the feasibility of sensor-based measurement to drive feedback and learning across individual, team, unit, and organizational levels, and future research is needed to refine methods, technologies, theory, and analytical strategies.



Point process modelling for directed interaction networks

A model is introduced for treating directed interactions as a multivariate point process: a Cox multiplicative intensity model using covariates that depend on the history of the process, and consistency and asymptotic normality are proved.

Attribute Fusion in a Latent Process Model for Time Series of Graphs

Inferential performance in mathematically tractable first-order and second-order approximation models does provide guidance for methodological choices applicable to the exact model, and it is tentatively concluded that approximation model investigations have some bearing on real data applications.

Inferring relevant social networks from interpersonal communication

It is found that prediction accuracy is maximized over a non-trivial range of thresholds corresponding to 5-10 reciprocated emails per year and that for any prediction task, choosing the optimal value of the threshold yields a sizable boost in accuracy over naive choices.

A latent process model for time series of attributed random graphs

A latent process model for time series of attributed random graphs for characterizing multiple modes of association among a collection of actors over time and it is demonstrated that the analysis through approximation can provide valuable information regarding inference properties.

A Poisson model for random multigraphs

The theory is applied to real data on neuronal connections, interacting genes in radiation hybrids, interacting proteins in a literature curated database, and letter and word pairs in seven Shaskespearean plays.

Bayesian anomaly detection methods for social networks

The first stage uses simple, conjugate Bayesian models for discrete time counting processes to track the pairwise links of all nodes in the graph to assess normality of behavior, and the second stage applies standard network inference tools on a greatly reduced subset of potentially anomalous nodes.

How to search a social network

Computational Methods for Dynamic Graphs

A data structure and an updating scheme are introduced that captures, in an approximate sense, the graph and its evolution through time and can be enhanced to form what the authors call communities of interest (COI).

Reconstruction of missing data in social networks based on temporal patterns of interactions

We discuss a mathematical framework based on a self-exciting point process aimed at analyzing temporal patterns in the series of interaction events between agents in a social network. We then develop

On Tensors, Sparsity, and Nonnegative Factorizations

This paper proposes that the random variation is best described via a Poisson distribution, which better describes the zeros observed in the data as compared to the typical assumption of a Gaussian distribution, and presents a new algorithm for Poisson tensor factorization called CANDECOMP--PARAFAC alternating Poisson regression (CP-APR), based on a majorization-minimization approach.