Anomalous Networks Research Articles

Multi-view discriminative edge heterophily contrastive learning network for attributed graph anomaly detection

Attributed graph anomaly detection aims to identify abnormal nodes that significantly differ from most nodes in terms of their attribute or structure. Recent graph contrastive learning methods, which follow an augmenting-contrasting learning scheme, have shown promising results. However, these methods still have two key limitations. Firstly, the contrastive views generated through random perturbation may not conform to abnormal patterns. Secondly, the inability to effectively discriminate heterophilic edges in anomalous networks can adversely impact the detection performance. To this end, we propose a Multi-view discriminative Edge heterophiLy cOntrastive learning Network (MELON) for attributed graph anomaly detection. Our approach addresses the aforementioned limitations by integrating human knowledge of different anomaly patterns for data augmentation and designing an edge discriminator to identify edge homophily/heterophily in an unsupervised manner. In addition, a dual-channel encoder is also designed to capture representative representations of the nodes from the discriminated homophilic/heterophilic edges. We then use a decoder to reconstruct the original network from the node representations learned by the dual-channel encoder and rank nodes according to their reconstruction error as the anomaly metric. Extensive experiments on four public benchmark datasets demonstrate that our proposed method outperforms state-of-the-art baselines.

Anomalous networks under the multispecies coalescent: theory and prevalence.

Reticulations in a phylogenetic network represent processes such as gene flow, admixture, recombination and hybrid speciation. Extending definitions from the tree setting, an anomalous network is one in which some unrooted tree topology displayed in the network appears in gene trees with a lower frequency than a tree not displayed in the network. We investigate anomalous networks under the Network Multispecies Coalescent Model with possible correlated inheritance at reticulations. Focusing on subsets of 4 taxa, we describe a new algorithm to calculate quartet concordance factors on networks of any level, faster than previous algorithms because of its focus on 4 taxa. We then study topological properties required for a 4-taxon network to be anomalous, uncovering the key role of [Formula: see text]-cycles: cycles of 3 edges parent to a sister group of 2 taxa. Under the model of common inheritance, that is, when each gene tree coalesces within a species tree displayed in the network, we prove that 4-taxon networks are never anomalous. Under independent and various levels of correlated inheritance, we use simulations under realistic parameters to quantify the prevalence of anomalous 4-taxon networks, finding that truly anomalous networks are rare. At the same time, however, we find a significant fraction of networks close enough to the anomaly zone to appear anomalous, when considering the quartet concordance factors observed from a few hundred genes. These apparent anomalies may challenge network inference methods.

Detecting Anomalous Networks of Opioid Prescribers and Dispensers in Prescription Drug Data

The opioid overdose epidemic represents a serious public health crisis, with fatality rates rising considerably over the past several years. To help address the abuse of prescription opioids, state governments collect data on dispensed prescriptions, yet the use of these data is typically limited to manual searches. In this paper, we propose a novel graph-based framework for detecting anomalous opioid prescribing patterns in state Prescription Drug Monitoring Program (PDMP) data, which could aid governments in deterring opioid diversion and abuse. Specifically, we seek to identify connected networks of opioid prescribers and dispensers who engage in high-risk and possibly illicit activity. We develop and apply a novel extension of the Non-Parametric Heterogeneous Graph Scan (NPHGS) to two years of de-identified PDMP data from the state of Kansas, and find that NPHGS identifies subgraphs that are significantly more anomalous than those detected by other graph-based methods. NPHGS also reveals clusters of potentially illicit activity, which may strengthen state law enforcement and regulatory capabilities. Our paper is the first to demonstrate how prescription data can systematically identify anomalous opioid prescribers and dispensers, as well as illustrating the efficacy of a network-based approach. Additionally, our technical extensions to NPHGS offer both improved flexibility and graph density reduction, enabling the framework to be replicated across jurisdictions and extended to other problem domains.

Non‐parametric regression for networks

Network data are becoming increasingly available, and so there is a need to develop a suitable methodology for statistical analysis. Networks can be represented as graph Laplacian matrices, which are a type of manifold‐valued data. Our main objective is to estimate a regression curve from a sample of graph Laplacian matrices conditional on a set of Euclidean covariates, for example, in dynamic networks where the covariate is time. We develop an adapted Nadaraya–Watson estimator which has uniform weak consistency for estimation using Euclidean and power Euclidean metrics. We apply the methodology to the Enron email corpus to model smooth trends in monthly networks and highlight anomalous networks. Another motivating application is given in corpus linguistics, which explores trends in an author's writing style over time based on word co‐occurrence networks.

Glutamate Levels Correlate With Anomalous Brain Functional Networks in First-Episode Psychosis Participants

Glutamate Levels Correlate With Anomalous Brain Functional Networks in First-Episode Psychosis Participants

Detecting anomalous payments networks: A dimensionality-reduction approach

Abstract Anomaly-detection methods are aimed at identifying observations that deviate manifestly from what is expected. Such methods are usually run on low-dimensional data, such as time series data. However, the increasing importance of high-dimensional payments and exposure data for financial oversight requires methods for detecting anomalous networks. To detect an anomalous network, dimensionality reduction allows measuring of the extent to which the network's main connective features (i.e. the structure) deviate from those regarded as typical. The key to dimensionality-reduction methods is the ability to reconstruct data with an error; this reconstruction error serves as a yardstick for deviation from what is typical. Principal component analysis (PCA) is used as a dimensionality-reduction method, and a clustering algorithm is used to classify reconstruction errors as normal or anomalous. Based on data from Colombia's large-value payments system and a set of synthetic anomalous networks created through simulations of intraday payments, detecting anomalous payments networks is feasible and promising for financial-oversight purposes.

Anomaly Detection With Subgraph Search and Vertex Classification Preprocessing in Chung-Lu Random Networks

We study network anomaly detection problem with subgraph search and vertex classification techniques in Chung-Lu networks, where small anomalous networks are embedded in background networks. The general detectors traditionally utilize the spectral characteristics of whole networks to decide whether the anomalous networks exist or not. Moreover, many algorithms model the background networks with special random graphs, such as Erdős–Renyi random graphs. However, these methods may not achieve good detection performance because the spectral information of anomalous networks extracting from the relational data of the whole networks is subtle. Furthermore, the assumptions of the models limit the applications of the methods. In order to improve the detection capability of algorithms and relax the restriction of the background networks models, we develop two preprocessing approaches, referred to as subgraph search and vertex classification, and propose two detection algorithms, which can be applicable to the generalized network model (Chung-Lu random graph). By leveraging statistic features of priori data, the subgraph search method can obtain local network nodes that are more anomalous than the remainder. Moreover, the vertex classification is used to further distinguish the anomalous nodes from the local ones. And then, by using the relational data corresponding to the local nodes, the detection statistics are constructed to detect the anomalous network. Additionally, based on concentration of measure theory, the probability bounds of performance analysis are derived for the presented algorithms. Simulation examples are provided to illustrate that the proposed detectors can achieve better detection performance than the existing algorithms.

Abstract 1851: A murine model of von Hippel-Lindau associated retinal hemangioblastoma

Abstract Purpose: von Hippel-Lindau disease (VHL) is an autosomal dominant condition that features a constellation of cysts or highly vascularized tumors due to loss of heterozygosity (LOH) of the VHL gene, resulting in deregulated hypoxia-induced factors (HIFs). Clinical manifestations of VHL include hemangioblastomas in the central nervous system and retina. To date, no VHL animal model has reproduced retinal capillary hemangioblastomas (RCH), the hallmark lesion of ocular VHL. The RCHs may arise from arrested hemangioblast progenitor cells, with VHL LOH in the tumor stroma. The stem cell leukemia (SCL) gene encodes a basic helix-loop-helix (bHLH) transcription factor, a critical regulator of hematopoiesis and vasculogenesis. We aim to generate a murine model of VHL-associated RCH by conditionally inactivating Vhl in a hemangioblast progenitor population with the angioblast-specific Cre line, (SCL) Cre-ERt2. Methods: We established a genetic mouse model in which the inducible Vhl gene is specifically “knocked-out” in angioblast cells using the stem cell leukemia (SCL) Cre-ERt2 transgenic mice. Two-week-old SCL-CreERt2;Vhl-F/F mice were administrated tamoxifen (2mg/40g body weight) for 5 consecutive days to activate the Cre recombinase that induces Vhl deletion in angioblast-derived cells. Fundoscopy was done monthly to detect retinal lesions. Fluorescein angiography (FA) was performed on affected mice. All retinas were analyzed by histology at 4 months of age. Genome typing of the Vhl conditional KO allele was conducted in retinal lesions using microdissection, nest-PCR and Sanger sequencing. Results: About 64% (18/28) of the transgenic mice exhibited various retinal vascular defects following induction. Affected mice demonstrated retinal vascular lesions that were variably associated with prominent vessels, anomalous capillary networks, hemorrhage, exudates, and localized fibrosis. FA revealed vascular leakage from the lesions. Histological analyses showed RCH-like lesions of tortuous, dilated vessels surrounded by “tumorlet” cells, isolated foamy stromal cells, and glia, classically found in RCH. Vhl LOH was detected in the tumor-like area as verified by sequencing. Conclusion: This is the first demonstration of VHL-associated RCH in a transgenic mouse model. This model may be useful for studying RCH pathogenesis, including HIF-dependent and HIF-independent pathways, and for testing potential therapies. Citation Format: Herui Wang, Qi Zhang, Lijin Dong, Liliana Guedez, Stanley Park, Yujuan Wang, Shida Chen, Chun Gao, Wai T. Wong, Henry Wiley, Emily Chew, Chi-Chao Chan, Zhengping Zhuang. A murine model of von Hippel-Lindau associated retinal hemangioblastoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 1851. doi:10.1158/1538-7445.AM2017-1851

Graph Analysis for Detecting Fraud, Waste, and Abuse in Healthcare Data

Detection of fraud, waste, and abuse (FWA) is an important yet difficult problem. In this paper, we describe a system to detect suspicious activities in large healthcare claims datasets. Each healthcare dataset is viewed as a heterogeneous network of patients, doctors, pharmacies, and other entities. These networks can be large, with millions of patients, hundreds of thousands of doctors, and tens of thousands of pharmacies, for example. Graph analysis techniques are developed to find suspicious individuals, suspicious relationships between individuals, unusual changes over time, unusual geospatial dispersion, and anomalous networks within the overall graph structure. The system has been deployed on multiple sites and data sets, both government and commercial, to facilitate the work of FWA investigation analysts.

Indicios de antecedencia en la red fluvial del sureste de Galicia

The present work offers a preliminary analysis of the river network with tertiary ante­ cedents of the Galician SE. The geomorphological observations of the area are centred on the cartography of erosive terraces, scrolled canals, hanging meanders, elbows of cap­ ture, and anomalous networks. For its interpretation we interface them with the thalwegs of the principal courses and the Alpine fractures charted by other authors. We propose a chronology for the identified fluvial processes, a chronology which points to a greater river network than has been estimated so far. Of these seven identified tendencies three presented a regional entity (ENE-WSW, NE-SW, N-S) and four local (NW-SE, SW-NE, SE-NW, S-N). The principal character of the paleo-network ENE-WSW (river Sil) is confirmed. And as a hypothesis we propose, for the Sierra de Queixa-San Mamede, the character of positive paleo-relief of Mesozoic inheritance. This relief has undergone several processes of isostatic raising and also tectonic during the Alpine Mountain-buil­ ding. These raisings have caused the superimposition of captures in the SE spurs of the Sierra de Queixa-San Mamede.