On fast and scalable recurring link’s prediction in evolving multi-graph streams

2020 ◽  
Vol 8 (S1) ◽  
pp. S65-S81 ◽  
Author(s):  
Shazia Tabassum ◽  
Bruno Veloso ◽  
João Gama
Keyword(s):  
Real World ◽  
Link Prediction ◽  
State Of The Art ◽  
Occurrence Frequency ◽  
Graph Topology ◽  
Empirical Results ◽  
Graph Streams ◽  
Dynamic Sampling ◽  
Fast Memory ◽  
Over Time

AbstractThe link prediction task has found numerous applications in real-world scenarios. However, in most of the cases like interactions, purchases, mobility, etc., links can re-occur again and again across time. As a result, the data being generated is excessively large to handle, associated with the complexity and sparsity of networks. Therefore, we propose a very fast, memory-less, and dynamic sampling-based method for predicting recurring links for a successive future point in time. This method works by biasing the links exponentially based on their time of occurrence, frequency, and stability. To evaluate the efficiency of our method, we carried out rigorous experiments with massive real-world graph streams. Our empirical results show that the proposed method outperforms the state-of-the-art method for recurring links prediction. Additionally, we also empirically analyzed the evolution of links with the perspective of multi-graph topology and their recurrence probability over time.

scholarly journals Lifelong Spectral Clustering

2020 ◽  
Vol 34 (04) ◽  
pp. 5867-5874
Author(s):  
Gan Sun ◽  
Yang Cong ◽  
Qianqian Wang ◽  
Jun Li ◽  
Yun Fu
Keyword(s):  
Machine Learning ◽  
Real World ◽  
Spectral Clustering ◽  
State Of The Art ◽  
Clustering Algorithms ◽  
Orthogonal Basis ◽  
Learning Framework ◽  
The Past ◽  
Benchmark Datasets ◽  
Over Time

In the past decades, spectral clustering (SC) has become one of the most effective clustering algorithms. However, most previous studies focus on spectral clustering tasks with a fixed task set, which cannot incorporate with a new spectral clustering task without accessing to previously learned tasks. In this paper, we aim to explore the problem of spectral clustering in a lifelong machine learning framework, i.e., Lifelong Spectral Clustering (L2SC). Its goal is to efficiently learn a model for a new spectral clustering task by selectively transferring previously accumulated experience from knowledge library. Specifically, the knowledge library of L2SC contains two components: 1) orthogonal basis library: capturing latent cluster centers among the clusters in each pair of tasks; 2) feature embedding library: embedding the feature manifold information shared among multiple related tasks. As a new spectral clustering task arrives, L2SC firstly transfers knowledge from both basis library and feature library to obtain encoding matrix, and further redefines the library base over time to maximize performance across all the clustering tasks. Meanwhile, a general online update formulation is derived to alternatively update the basis library and feature library. Finally, the empirical experiments on several real-world benchmark datasets demonstrate that our L2SC model can effectively improve the clustering performance when comparing with other state-of-the-art spectral clustering algorithms.

scholarly journals Stacking models for nearly optimal link prediction in complex networks

2020 ◽  
Vol 117 (38) ◽  
pp. 23393-23400 ◽  
Author(s):  
Amir Ghasemian ◽  
Homa Hosseinmardi ◽  
Aram Galstyan ◽  
Edoardo M. Airoldi ◽  
Aaron Clauset
Keyword(s):  
Social Networks ◽  
Data Collection ◽  
Real World ◽  
Link Prediction ◽  
State Of The Art ◽  
Network Data ◽  
Prediction Errors ◽  
Partially Observed ◽  
Speed Up ◽  
Large Corpus

Most real-world networks are incompletely observed. Algorithms that can accurately predict which links are missing can dramatically speed up network data collection and improve network model validation. Many algorithms now exist for predicting missing links, given a partially observed network, but it has remained unknown whether a single best predictor exists, how link predictability varies across methods and networks from different domains, and how close to optimality current methods are. We answer these questions by systematically evaluating 203 individual link predictor algorithms, representing three popular families of methods, applied to a large corpus of 550 structurally diverse networks from six scientific domains. We first show that individual algorithms exhibit a broad diversity of prediction errors, such that no one predictor or family is best, or worst, across all realistic inputs. We then exploit this diversity using network-based metalearning to construct a series of “stacked” models that combine predictors into a single algorithm. Applied to a broad range of synthetic networks, for which we may analytically calculate optimal performance, these stacked models achieve optimal or nearly optimal levels of accuracy. Applied to real-world networks, stacked models are superior, but their accuracy varies strongly by domain, suggesting that link prediction may be fundamentally easier in social networks than in biological or technological networks. These results indicate that the state of the art for link prediction comes from combining individual algorithms, which can achieve nearly optimal predictions. We close with a brief discussion of limitations and opportunities for further improvements.

scholarly journals Finding Missing Links in Complex Networks: A Multiple-Attribute Decision-Making Method

Complexity ◽  
2018 ◽  
Vol 2018 ◽  
pp. 1-16 ◽  
Author(s):  
Longjie Li ◽  
Shenshen Bai ◽  
Mingwei Leng ◽  
Lu Wang ◽  
Xiaoyun Chen
Keyword(s):  
Decision Making ◽  
Complex Networks ◽  
Similarity Measure ◽  
Real World ◽  
Link Prediction ◽  
State Of The Art ◽  
Multiple Attribute Decision Making ◽  
Ideal Solution ◽  
Multiple Attribute ◽  
Novel Method

Link prediction, which aims to forecast potential or missing links in a complex network based on currently observed information, has drawn growing attention from researchers. To date, a host of similarity-based methods have been put forward. Usually, one method harbors the idea that one similarity measure is applicable to various networks, and thus has performance fluctuation on different networks. In this paper, we propose a novel method to solve this issue by regarding link prediction as a multiple-attribute decision-making (MADM) problem. In the proposed method, we consider RA, LP, and CAR indices as the multiattribute for node pairs. The technique for order performance by similarity to ideal solution (TOPSIS) is adopted to aggregate the multiattribute and rank node pairs. The proposed method is not limited to only one similarity measure, but takes separate measures into account, since different networks may have different topological structures. Experimental results on 10 real-world networks manifest that the proposed method is superior in comparison to state-of-the-art methods.

scholarly journals A Generative Model for Dynamic Networks with Applications

2019 ◽  
Vol 33 ◽  
pp. 7842-7849
Author(s):  
Shubham Gupta ◽  
Gaurav Sharma ◽  
Ambedkar Dukkipati
Keyword(s):  
Neural Network ◽  
Social Networks ◽  
Real World ◽  
Link Prediction ◽  
Temporal Dynamics ◽  
Dynamic Networks ◽  
Collaboration Networks ◽  
Proposed Model ◽  
Latent Space ◽  
Over Time

Networks observed in real world like social networks, collaboration networks etc., exhibit temporal dynamics, i.e. nodes and edges appear and/or disappear over time. In this paper, we propose a generative, latent space based, statistical model for such networks (called dynamic networks). We consider the case where the number of nodes is fixed, but the presence of edges can vary over time. Our model allows the number of communities in the network to be different at different time steps. We use a neural network based methodology to perform approximate inference in the proposed model and its simplified version. Experiments done on synthetic and real world networks for the task of community detection and link prediction demonstrate the utility and effectiveness of our model as compared to other similar existing approaches.

The New Economic Populism and the Future of Inequality in the United States

2017 ◽  
Author(s):  
William W. Franko ◽  
Christopher Witko
Keyword(s):  
United States ◽  
Federal Government ◽  
Labor Unions ◽  
Republican Party ◽  
The United States ◽  
The Public ◽  
Empirical Results ◽  
Income Disparities ◽  
The Face ◽  
Over Time

The authors conclude the book by recapping their arguments and empirical results, and discussing the possibilities for the “new economic populism” to promote egalitarian economic outcomes in the face of continuing gridlock and the dominance of Washington, DC’s policymaking institutions by business and the wealthy, and a conservative Republican Party. Many states are actually addressing inequality now, and these policies are working. Admittedly, many states also continue to embrace the policies that have contributed to growing inequality, such as tax cuts for the wealthy or attempting to weaken labor unions. But as the public grows more concerned about inequality, the authors argue, policies that help to address these income disparities will become more popular, and policies that exacerbate inequality will become less so. Over time, if history is a guide, more egalitarian policies will spread across the states, and ultimately to the federal government.

scholarly journals Higher-order temporal network effects through triplet evolution

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Qing Yao ◽  
Bingsheng Chen ◽  
Tim S. Evans ◽  
Kim Christensen
Keyword(s):  
Real World ◽  
Link Prediction ◽  
Higher Order ◽  
Prediction Algorithm ◽  
Interaction Patterns ◽  
Temporal Networks ◽  
World Systems ◽  
Order Interaction ◽  
Space And Time ◽  
Pairwise Interactions

AbstractWe study the evolution of networks through ‘triplets’—three-node graphlets. We develop a method to compute a transition matrix to describe the evolution of triplets in temporal networks. To identify the importance of higher-order interactions in the evolution of networks, we compare both artificial and real-world data to a model based on pairwise interactions only. The significant differences between the computed matrix and the calculated matrix from the fitted parameters demonstrate that non-pairwise interactions exist for various real-world systems in space and time, such as our data sets. Furthermore, this also reveals that different patterns of higher-order interaction are involved in different real-world situations. To test our approach, we then use these transition matrices as the basis of a link prediction algorithm. We investigate our algorithm’s performance on four temporal networks, comparing our approach against ten other link prediction methods. Our results show that higher-order interactions in both space and time play a crucial role in the evolution of networks as we find our method, along with two other methods based on non-local interactions, give the best overall performance. The results also confirm the concept that the higher-order interaction patterns, i.e., triplet dynamics, can help us understand and predict the evolution of different real-world systems.

scholarly journals Real World Performance Evaluation of Transcatheter Aortic Valve Implantation

2021 ◽  
Vol 10 (9) ◽  
pp. 1890
Author(s):  
Gabriele Pesarini ◽  
Gabriele Venturi ◽  
Domenico Tavella ◽  
Leonardo Gottin ◽  
Mattia Lunardi ◽  
...  
Keyword(s):  
Aortic Valve ◽  
Real World ◽  
Baseline Risk ◽  
Risk Scores ◽  
World Population ◽  
Control Analysis ◽  
Transcatheter Aortic Valve ◽  
Procedural Control ◽  
Risk Patients ◽  
Over Time

Background: The aim of this research is to describe the performance over time of transcatheter aortic valve implantations (TAVIs) in a high-volume center with a contemporary, real-world population. Methods: Patients referred for TAVIs at the University Hospital of Verona were prospectively enrolled. By cumulative sum failures analysis (CUSUM), procedural-control curves for standardized combined endpoints—as defined by the Valve Academic Research Consortium-2 (VARC-2)—were calculated and analyzed over time. Acceptable and unacceptable limits were derived from recent studies on TAVI in intermediate and low-risk patients to fit the higher required standards for current indications. Results: A total of 910 patients were included. Baseline risk scores significantly reduced over time. Complete procedural control was obtained after approximately 125 and 190 cases for device success and early safety standardized combined endpoints, respectively. High risk patients (STS ≥ 8) had poorer outcomes, especially in terms of VARC-2 clinical efficacy, and required a higher case load to maintain in-control and proficient procedures. Clinically relevant single endpoints were all influenced by operator’s experience as well. Conclusions: Quality-control analysis for contemporary TAVI interventions based on standardized endpoints suggests the need for relevant operator’s experience to achieve and maintain optimal clinical results, especially in higher-risk subjects.

scholarly journals TransET: Knowledge Graph Embedding with Entity Types

Electronics ◽  
2021 ◽  
Vol 10 (12) ◽  
pp. 1407
Author(s):  
Peng Wang ◽  
Jing Zhou ◽  
Yuzhang Liu ◽  
Xingchen Zhou
Keyword(s):  
Link Prediction ◽  
State Of The Art ◽  
Score Function ◽  
Graph Embedding ◽  
Vector Spaces ◽  
Knowledge Graph ◽  
Semantic Features ◽  
Knowledge Graphs ◽  
Real World Datasets ◽  
Low Dimensional

Knowledge graph embedding aims to embed entities and relations into low-dimensional vector spaces. Most existing methods only focus on triple facts in knowledge graphs. In addition, models based on translation or distance measurement cannot fully represent complex relations. As well-constructed prior knowledge, entity types can be employed to learn the representations of entities and relations. In this paper, we propose a novel knowledge graph embedding model named TransET, which takes advantage of entity types to learn more semantic features. More specifically, circle convolution based on the embeddings of entity and entity types is utilized to map head entity and tail entity to type-specific representations, then translation-based score function is used to learn the presentation triples. We evaluated our model on real-world datasets with two benchmark tasks of link prediction and triple classification. Experimental results demonstrate that it outperforms state-of-the-art models in most cases.

scholarly journals POS0619 MODELLING OF DISEASE ACTIVITY IN PATIENTS WITH INFLAMMATORY ARTHROPATHIES TREATED WITH ETANERCEPT ORIGINATOR OR BIOSIMILAR AS FIRST-LINE BIOLOGIC IN AN AUSTRALIAN REAL-WORLD DATASET

2021 ◽  
Vol 80 (Suppl 1) ◽  
pp. 547.1-547
Author(s):  
C. Deakin ◽  
G. Littlejohn ◽  
H. Griffiths ◽  
T. Smith ◽  
C. Osullivan ◽  
...  
Keyword(s):  
Disease Activity ◽  
Clinical Data ◽  
Real World ◽  
Rank Test ◽  
P Value ◽  
Continuous Variables ◽  
Linear Quadratic ◽  
First Line ◽  
Modest Improvement ◽  
Over Time

Background:The availability of biosimilars as non-proprietary versions of established biologic disease-modifying anti-rheumatic drugs (bDMARDs) is enabling greater access for patients with rheumatic diseases to effective medications at a lower cost. Since April 2017 both the originator and a biosimilar for etanercept (trade names Enbrel and Brenzys, respectively) have been available for use in Australia.Objectives:[1]To model effectiveness of etanercept originator or biosimilar in reducing Disease Activity Score 28-joint count C reactive protein (DAS28CRP) in patients with rheumatoid arthritis (RA), psoriatic arthritis (PsA) or ankylosing spondylitis (AS) treated with either drug as first-line bDMARD[2]To describe persistence on etanercept originator or biosimilar as first-line bDMARD in patients with RA, PsA or ASMethods:Clinical data were obtained from the Optimising Patient outcomes in Australian rheumatoLogy (OPAL) dataset, derived from electronic medical records. Eligible patients with RA, PsA or AS who initiated etanercept originator (n=856) or biosimilar (n=477) as first-line bDMARD between 1 April 2017 and 31 December 2020 were identified. Propensity score matching was performed to select patients on originator (n=230) or biosimilar (n=136) with similar characteristics in terms of diagnosis, disease duration, joint count, age, sex and concomitant medications. Data on clinical outcomes were recorded at 3 months after baseline, and then at 6-monthly intervals. Outcomes data that were missing at a recorded visit were imputed.Effectiveness of the originator, relative to the biosimilar, for reducing DAS28CRP over time was modelled in the matched population using linear mixed models with both random intercepts and slopes to allow for individual heterogeneity, and weighting of individuals by inverse probability of treatment weights to ensure comparability between treatment groups. Time was modelled as a combination of linear, quadratic and cubic continuous variables.Persistence on the originator or biosimilar was analysed using survival analysis (log-rank test).Results:Reduction in DAS28CRP was associated with both time and etanercept originator treatment (Table 1). The conditional R-squared for the model was 0.31. The average predicted DAS28CRP at baseline, 3 months, 6 months, 9 months and 12 months were 4.0 and 4.4, 3.1 and 3.4, 2.6 and 2.8, 2.3 and 2.6, and 2.2 and 2.4 for the originator and biosimilar, respectively, indicating a clinically meaningful effect of time for patients on either drug and an additional modest improvement for patients on the originator.Median time to 50% of patients stopping treatment was 25.5 months for the originator and 24.1 months for the biosimilar (p=0.53). An adverse event was the reason for discontinuing treatment in 33 patients (14.5%) on the originator and 18 patients (12.9%) on the biosimilar.Conclusion:Analysis using a large national real-world dataset showed treatment with either the etanercept originator or the biosimilar was associated with a reduction in DAS28CRP over time, with the originator being associated with a further modest reduction in DAS28CRP that was not clinically significant. Persistence on treatment was not different between the two drugs.Table 1.Respondent characteristics.Fixed EffectEstimate95% Confidence Intervalp-valueTime (linear)0.900.89, 0.911.5e-63Time (quadratic)1.011.00, 1.011.3e-33Time (cubic)1.001.00, 1.007.1e-23Originator0.910.86, 0.960.0013Acknowledgements:The authors acknowledge the members of OPAL Rheumatology Ltd and their patients for providing clinical data for this study, and Software4Specialists Pty Ltd for providing the Audit4 platform.Supported in part by a research grant from Investigator-Initiated Studies Program of Merck & Co Inc, Kenilworth, NJ, USA. The opinions expressed in this paper are those of the authors and do not necessarily represent those of Merck & Co Inc, Kenilworth, NJ, USA.Disclosure of Interests:Claire Deakin: None declared, Geoff Littlejohn Consultant of: Over the last 5 years Geoffrey Littlejohn has received educational grants and consulting fees from AbbVie, Bristol Myers Squibb, Eli Lilly, Gilead, Novartis, Pfizer, Janssen, Sandoz, Sanofi and Seqirus., Hedley Griffiths Consultant of: AbbVie, Gilead, Novartis and Lilly., Tegan Smith: None declared, Catherine OSullivan: None declared, Paul Bird Speakers bureau: Eli Lilly, abbvie, pfizer, BMS, UCB, Gilead, Novartis

scholarly journals C3VFC: A Method for Tracing and Quantification of Microglia in 3D Temporal Images

2021 ◽  
Vol 11 (13) ◽  
pp. 6078
Author(s):  
Tiffany T. Ly ◽  
Jie Wang ◽  
Kanchan Bisht ◽  
Ukpong Eyo ◽  
Scott T. Acton
Keyword(s):  
Vector Field ◽  
Critical Points ◽  
State Of The Art ◽  
Temporal Data ◽  
Entire Structure ◽  
Tracing Algorithm ◽  
Vector Field Convolution ◽  
Multiple Cells ◽  
Tracing Method ◽  
Over Time

Automatic glia reconstruction is essential for the dynamic analysis of microglia motility and morphology, notably so in research on neurodegenerative diseases. In this paper, we propose an automatic 3D tracing algorithm called C3VFC that uses vector field convolution to find the critical points along the centerline of an object and trace paths that traverse back to the soma of every cell in an image. The solution provides detection and labeling of multiple cells in an image over time, leading to multi-object reconstruction. The reconstruction results can be used to extract bioinformatics from temporal data in different settings. The C3VFC reconstruction results found up to a 53% improvement on the next best performing state-of-the-art tracing method. C3VFC achieved the highest accuracy scores, in relation to the baseline results, in four of the five different measures: Entire structure average, the average bi-directional entire structure average, the different structure average, and the percentage of different structures.

Sign in / Sign up

Export Citation Format

Share Document