Structural network inference from time-series data using a generative model and transfer entropy

Abstract

In this paper we develop a novel framework for inferring a generative model of network structure representing the causal relations between data for a set of objects characterized in terms of time series. To do this we make use of transfer entropy as a means of inferring directed information transfer between the time-series data. Transfer entropy allows us to infer directed edges representing the causal relations between pairs of time series, and has thus been used to infer directed graph representations of causal networks for time-series data. We use the expectation maximization algorithm to learn a generative model which captures variations in the causal network over time. We conduct experiments on fMRI brain connectivity data for subjects in different stages of the development of Alzheimer's disease (AD). Here we use the technique to learn class exemplars for different stages in the development of the disease, together with a normal control class, and demonstrate its utility in both graph multi-class and binary classifications. These experiments are showing the effectiveness of our proposed framework when the amounts of training data are relatively small.

Improving best route using intelligent Ad Hoc system

“…Number of barges: 45 Number of generation: 1000 Population size: 1000 Mutation Probability: 0.08 The red barge is the actual barge that selects the best course (3,4,5,6,7,8,9,10,11,13,14,15,16,17,19,20 21,22,23,24,25,27,28,29,31,32,34,38,40,42,44) mixed with seven black barge that form the secondary towers (Emergency ( are (1,2,12,26,18,41,43 45) and there are five barge which are stopped its colored green (31,37,35,33,39) as shown in figure 6 and cost evaluation with number of barges 45 shown in figure 7. The red barge is the actual barge that selects the best course (1, 3, 4, 5,6 ,7, 8, 9,10, 11,13, 14, 15, 16, 20, 21, 23, 24, 25, 27, 30, 32, 36, 38, 40, 41, 42, 43, 44, 45, 46, 47,48 ,49 (1,2,3,4,5,9,12,13,14,16,17,18,19,21,21,22,25,26,27,28,29,30,32,34,36,38,40,41,42,43,44,45) mixed with eight black barge that form the secondary towers (Emergency) are (6,7,8,10,11,15,23,24) Generation number: 200 mutation probability: 0.08 The red barge is the actual barge that selects the best course (1,2,3, ...…”

“…It only needs transceiver and equipped without infrastructure, and it can do its communication with other nodes. The nodes can connect with other nodes that exist in another area in the infrastructure based wireless network and transferring the data [9]. The ad hoc networks are not effective if compared with a large network such as infrastructure network wherever; the node has a limited range if the nodes ranged are combined, that would produce large data transmission area [10].…”

Network inference via process motifs for lagged correlation in linear stochastic processes

“…Researchers from diverse disciplines have proposed many pairwise edge measures (PEMs) for the inference of causal relationships between pairs of variables [7,8] and various network-inference problems [4,[9][10][11][12][13][14][15][16][17][18][19][20][21]. Possibly the simplest PEMs are correlation [10][11][12] and lagged correlation [13].…”

“…In neuroscience, it is common to refer to networks inferred via correlations or lagged correlations in time-series data as networks of "functional connectivity" (as opposed to "structural connectivity") to avoid a claim that correlations or lagged correlations indicate causal relationships [4]. Many PEMs are based on information-theoretic considerations [14][15][16][17][18][19][20][21]. These PEMs include mutual information [14,15], transfer entropy [16][17][18], partial information decomposition [19], unique information [22], and variants thereof [20,21].…”

“…Many PEMs are based on information-theoretic considerations [14][15][16][17][18][19][20][21]. These PEMs include mutual information [14,15], transfer entropy [16][17][18], partial information decomposition [19], unique information [22], and variants thereof [20,21]. A very popular PEM is linear Granger causality, which is a measure of edge likelihood based on fitting a VAR model [23,24].…”