We consider the problem of analyzing timestamped relational events between a set of entities, such as messages between users of an on-line social network. Such data are often analyzed using static or discrete-time network models, which discard a significant amount of information by aggregating events over time to form network snapshots. In this paper, we introduce a block point process model (BPPM) for continuous-time event-based dynamic networks. The BPPM is inspired by the well-known stochastic block model (SBM) for static networks. We show that networks generated by the BPPM follow an SBM in the limit of a growing number of nodes. We use this property to develop principled and efficient local search and variational inference procedures initialized by regularized spectral clustering. We fit BPPMs with exponential Hawkes processes to analyze several real network data sets, including a Facebook wall post network with over 3, 500 nodes and 130, 000 events.

In this paper, we investigate several metrics currently used for evaluating accuracies of dynamic link prediction methods and demonstrate why they can be misleading in many cases. We provide several recommendations on evaluating dynamic link prediction accuracy, including separation into two categories of evaluation. Finally we propose a unified metric to characterize link prediction accuracy effectively using a single number.

© 2015 Wiley Periodicals, Inc.The modeling of the physical and electrical characteristics of microstrip non-uniform transmission lines (NTLs) utilizing artificial neural networks (ANNs) is investigated. The fundamental equations and constraints for designing variable impedance transmission lines are first presented. Then, a proof-of-concept example of a compact non-uniform matching transformer and the counterpart modeled version is elaborated for source and load impedances Zs and Zl, respectively, at 0.5 GHz. For comparison purposes, weights and biases of the proposed ANN are established with three different training techniques; namely: backpropagation (BP), Quasi-Newton (QN), and conjugate gradient (CG); at which the ABCD matrix, impedance variations, input port matching (S11), and transmission parameter (S21) are set as benchmarks to examine the validity of the trained model. The concept is then extended to model a NTL ultrawideband (UWB) Wilkinson power divider (WPD) with three resistors for improved isolation. S-parameters derived from the trained ANN outputs are close to those obtained by the traditional time-consuming optimization procedure, and show input and output ports matching and isolation of below -10 dB, and acceptable values of transmission parameters over the 3.1 GHz to 10.6 GHz band. The resulting models outperform traditional optimizations in terms of simulation time and reserved resources with comparable accuracy

No abstract

No abstract