Graph pattern matching is typically defined in terms of subgraph isomorphism, which makes it an np-complete problem. Moreover, requires bijective functions, are often too restrictive to characterize patterns emerging applications. We propose a class graph patterns, edge denotes the connectivity data within predefined number hops. In addition, we define based on notion bounded simulation, extension simulation. show that with this revision, can be performed cubic-time, by providing such...

It is common to find graphs with millions of nodes and billions edges in, e.g., social networks. Queries on such are often prohibitively expensive. These motivate us propose query preserving graph compression, compress relative a class Λ queries users' choice. We compute small Gr from G that (a) for any Q Ε Q, Q(G) = Q'(Gr), where Q' can be efficiently computed Q; (b) algorithm computing directly applied evaluating as is. That is, while we cannot lower the complexity queries, reduce data...

It is increasingly common to find graphs in which edges bear different types, indicating a variety of relationships. For such we propose class reachability queries and graph patterns, an edge specified with regular expression certain form, expressing the connectivity data via various types. In addition, define pattern matching based on revised notion simulation. On emerging applications as social networks, show that these are capable finding more sensible information than their traditional...

It is increasingly common to find real-life data represented as networks of labeled, heterogeneous entities. To query these networks, one often needs identify the matches a given graph in (typically large) network modeled target graph. Due noise and lack fixed schema graph, can substantially differ from its both structure node labels, thus bringing challenges querying tasks. In this paper, we propose NeMa (Network Match), neighborhood-based subgraph matching technique for networks. (1)...

-With the advent of phasor measurement units (PMUs), high resolution synchronized measurements enables real time system monitoring and control. PMUs transmit data to local controllers in substations concentrators for wide control application center. They provide real-time critical power applications such as remedial action schemes, oscillation detection, state estimation. The quality from is smart grid applications. Several methods are developed detect anomalies series data, tailored PMU...

Graph pattern matching has been widely used in e.g., social data analysis. A number of algorithms have developed that, given a graph Q and G , compute the set M(Q,G) matches . However, these often return an excessive matches, are expensive on large real-life graphs. Moreover, practice many queries to find specific node, rather than entire This paper studies top- k matching. (1) We revise defined terms simulation, by supporting designated output node u o Given it is those nodes that match...

Graph pattern matching has become a routine process in emerging applications such as social networks. In practice data graph is typically large, and frequently updated with small changes. It often prohibitively expensive to recompute matches from scratch via batch algorithms when the updated. With this comes need for incremental that compute changes response updates, minimize unnecessary recomputation. This paper investigates defined terms of simulation, bounded simulation subgraph...

Graph pattern matching is commonly used in a variety of emerging applications such as social network analysis. These highlight the need for studying following two issues. First, graph traditionally defined terms subgraph isomorphism or simulation. notions, however, often impose too strong topological constraint on graphs to identify meaningful matches. Second, practice typically large, and frequently updated with small changes. It prohibitively expensive recompute matches starting from...

We propose graph-pattern association rules (GPARs) for social media marketing. Extending item-sets, GPARs help us discover regularities between entities in graphs, and identify potential customers by exploring influence. study the problem of discovering top- k diversified GPARs. While this is NP-hard, we develop a parallel algorithm with accuracy bound. also identifying it provide scalable that guarantees polynomial speedup over sequential algorithms increase processors. Using real-life...

We propose a class of functional dependencies for graphs, referred to as GFDs. GFDs capture both attribute-value and topological structures entities, subsume conditional (CFDs) special case. show that the satisfiability implication problems are coNP-complete NP-complete, respectively, no worse than their CFD counterparts. also validation problem is coNP-complete. Despite intractability, we develop parallel scalable algorithms catching violations in large-scale graphs. Using real-life...

Querying heterogeneous and large-scale knowledge graphs is expensive. This paper studies a graph summarization framework to facilitate search. (1) We introduce class of reduced summaries. Characterized by approximate pattern matching, these summaries are capable summarizing entities in terms their neighborhood similarity up certain hop, using small informative patterns. (2) study diversified problem. Given graph, it discover top-k that maximize bi-criteria function, characterized both...

Abstract The downhole monitoring of strain using Fiber Optics (FO) can reveal unique information about the propagation and geometry hydraulic fractures between nearby wells during stimulation production. This work aims at creating a catalogue commonly observed strain-rate signals captured in not yet stimulated observation well equipped with either permanently or temporarily installed FO cable. is result an informal collaboration experience users from academia, service providers, consulting...

In a variety of emerging applications one needs to decide whether graph G matches another p , i.e. has topological structure similar that . The traditional notions homomorphism and isomorphism often fall short capturing the structural similarity in these applications. This paper studies revisions notions, providing full treatment from complexity algorithms. (1) We propose p-homomorphism (p -hom) 1-1 -hom, which extend subgraph isomorphism, respectively, by mapping edges paths another,...

This paper presents GRAPE, a parallel system for graph computations. GRAPE differs from prior systems in its ability to parallelize existing sequential algorithms as whole. Underlying are simple programming model and principled approach, based on partial evaluation incremental computation. We show that can be "plugged into" with minor changes, get parallelized. As long the correct, their parallelization guarantees terminate correct answers under monotonic condition. Moreover, we MapReduce,...

Querying complex graph databases such as knowledge graphs is a challenging task for non-professional users. Due to their schemas and variational information descriptions, it becomes very hard users formulate query that can be properly processed by the existing systems. We argue user-friendly engine, must support various kinds of transformations synonym, abbreviation, ontology. Furthermore, derived results ranked in principled manner. In this paper, we introduce novel framework enabling...

Protection systems are one of the most critical components in transmission system and becoming more digital with ongoing automation. These prone to vulnerabilities/attacks, exploitation these vulnerabilities may cause major impacts on electric grid performance. Multiple alarms reported control center could be a result faults (expected operations) or failures protection (anomalies/ unexpected operation). Situational awareness gained through sensors such as phasor measurement unit (PMU) data...

Summary This paper presents a case study of fault reactivation and induced seismicity during multistage hydraulic fracturing in Sichuan Basin, China. The field microseismicity data delineate activated near the toe horizontal well. spatio-temporal characteristics indicate that seismic activity on first three stages is directly related to fluid injection, while after Stage 3, possibly due relaxation fault. fault-related events have larger magnitudes different frequency-magnitude compared...

With ongoing automation and digitization of the electric power system, several Phasor Measurement Units (PMUs) have been deployed for monitoring control. PMU data can multiple anomalies, many researchers in past concentrated on training machine/deep learning algorithms offline anomaly detection over (i.e., not real-time). These algorithms, when trained a sample rather than population dataset, fail to consider dynamic behavior grid real-time, resulting low accuracy. Considering (e.g., change...

Introduction With the rapid aging population, mental health of older adults is paid more and attention. Anxiety a common illness in adults. Therefore, study aimed to explore current situation anxiety its factors among elderly China. Methods Based on data from 2018 Chinese Longitudinal Healthy Longevity Survey (CLHLS), total 10,982 respondents aged 60 above were selected. Generalized Disorder (GAD-7) scale was used assess anxiety. Univariate multivariate analysis analyze influencing Random...

Given a graph query Q posed on knowledge G, top-k querying is to find k matches in G with the highest ranking score according function. Fast search graphs challenging as both traversal and similarity are expensive. Conventional typically based threshold algorithm (TA), which can no long fit demand new setting. This work proposes STAR, framework. It has two components: (a) fast for star queries, (b) an assembling general queries. The uses building block iteratively sweeps match lists...

There has recently been a lot of ongoing research in the areas fairness, bias and explainability machine learning (ML) models due to self-evident or regulatory requirements various ML applications. We make following observation: All these approaches require robust understanding relationship between data used train them. In this work, we introduce provenance tracking problem: fundamental idea is automatically track which columns dataset have derive features/labels an model. discuss challenges...

In recent years, a large number of photovoltaic (PV) systems have been added to the electrical grid as well installed off-grid systems. The trend suggests that deployment PV will continue rise in future. Thus, accurate forecasting performance is critical for reliability Due complex non-linear variability power output systems, non-trivial task. This affects stability and planning system network, can reduce uncertainty caused during operation. this work, we leverage spatial temporal coherence...

Generating explanations for graph neural networks (GNNs) has been studied to understand their behavior in analytical tasks such as classification. Existing approaches aim the overall results of GNNs rather than providing specific class labels interest, and may return explanation structures that are hard access, nor directly queryable.We propose GVEX, a novel paradigm generates Graph Views EXplanation. (1) We design two-tier structure called views. An view consists set patterns induced...