Analysing the degree distribution of real graphs by means of several probabilistic models

Mar 24, 20151 like869 views

The document analyzes the degree distribution of nodes in real-world networks using probabilistic models. It studies networks from datasets like Amazon, California roads, DBLP co-authorships, and others. Maximum likelihood and information criteria are used to determine the best fitting distributions, including Altmann, discrete Weibull, and MOEZipf. The analysis finds the MOEZipf distribution provides the best fit for most networks, followed by the discrete Weibull and Altmann distributions. Future work is proposed to integrate the best fitting distributions into a graph generation tool and analyze correlations between degree distribution and other network structures.

Analysing the degree distribution of real
graphs by means of several probabilistic
models
A. Duarte-L´opez, A. Prat-P´erez, M. P´erez-Casany
DAMA-UPC, Universitat Polit`ecnica de Catalunya
18th March, 2015

Introduction
Networks are complex structures in which the connections among
the nodes can exhibit complicated patterns.
Given a graph G with set of nodes V, the node degree is the number
of connections that a node has.
Objective: To determine the probability distribution of the de-
gree random variable of a real graph.
2/9

Networks
Diﬀerent real live networks from SNAP were studied.
Network Nodes Edges Type
Amazon 262111 1234876 Directed
CA roadnet 1965206 5533213 Undirected
DBLP Co-autorship 317080 1049865 Undirected
Live Journal 3997962 34681188 Undirected
NotreDame 325729 1497133 Directed
Patents 3774767 16518947 Directed
TX roadnet 1379917 3843319 Undirected
Wikipedia 2394385 5021409 Directed
Youtube 1134890 2987623 Undirected
3/9

Probabilistic Models
The non-negative integer probability distributions considered
are:
Uni-parametric Models Bi-parametric Models
Geometric Right-truncated Zipf
Poisson MOEZipf
Zipf Altmann
Negative Binomial
Discrete Weibull
All the models that contain the zero value in its domain were
truncated at 1.
4/9

Estimation and Goodness of the fit
Maximum likelihood method.
Given k = (k1, k2, ..., kN) a sample for the node degree we
maximize:
l(θ; k) =
N
i=1
log(pθ(ki))
Let M be the dimension of the parameter vector.
Akaike Information Criteria
AIC = −2l(ˆθ, K) + 2M
N
N − M − 1
Bayesian Information Criteria
BIC = −2l(ˆθ, k) + Mlog(N)
5/9

Results
1 10 100 1000 10000
110010000
Youtube (log−log scale)
Degree
Frequency
Observations
Altmann Distribution
Discrete Weibull Distribution
MOEZipf Distribution
Parameters
Distribution Parameters
Altmann ˆγ = 1.64; ˆδ = 0.0036
Discrete Weibull ˆv = 0.1424; ˆp = 0.0044
MOEZipf ˆγ = 2.089; ˆβ = 2.4101
Goodness of the ﬁt
Distribution AIC ∆AIC
Altmann 1688294.56 13402.66
Discrete Weibull 1676831.47 1939.57
MOEZipf 1674891.9 0
Distribution BIC ∆BIC
Altmann 1688316.23 13402.66
Discrete Weibull 1676853.14 1939.57
MOEZipf 1674913.57 0
6/9

Conclusions
The models given better ﬁts are:
MOEZipf model (54%)
Zero truncated discrete Weibull Model(38%)
Altmann Model (8%).
7/9

Future work
To integrate functions in the DataGen (LDBC project) that
allow generate random graphs in which their node degree
follows the MOEZipf, the zero truncated discrete Weibull
or the Altmann distributions.
To analyse the correlation between the degree distribution
and some other structural characteristics of the network
such as the clustering coeﬃcient, the degree assortativity,
etc.
8/9

Why should you come to our poster?
You will ﬁnd a relatively easy approach that allows you to
get more information about your network structure.
You can share with us your experience with respect the
node degree distribution.
Further potential collaborations.
9/9

This document examines tests conducted to measure the charge transfer efficiency (CTE) of a KAF-0402 CCD image sensor. An Fe55 radioactive source was used to generate single pixel events which could be tracked across the sensor to calculate CTE. A histogram of pixel values showed a clear spike at 500 counts corresponding to single events. A column histogram also showed a horizontal cluster of single events. By fitting a line and calculating the slope, the CTE was determined to be 0.99999, indicating minimal charge loss during transfer. The total charge loss was calculated to be approximately 1.3 electrons lost per 768 pixel transfers.

Forecasting Electricity Consumption Using Weather Data in Edge-Fog-Cloud Data...Juan Carlos Olivares Rojas

This document proposes an edge-fog-cloud computing architecture to forecast electricity consumption using weather data and machine learning models. The architecture implements linear regression models at the edge, fog, and cloud layers to improve forecasting accuracy. Results show forecast accuracy was improved with 91-94% of devices having better forecasts when models were executed across multiple layers. The architecture provides a feasible approach for electricity forecasting considering periodic offline execution of models across layers.

Interesting applications of graph theoryTech_MX

The document discusses graphs and their applications. It defines key graph terms like vertices, edges, directed/undirected graphs, paths, cycles, etc. It then describes algorithms for finding minimum spanning trees, Eulerian cycles, Hamiltonian paths, and approximations for the traveling salesman problem. Examples are provided to illustrate minimum spanning tree and Christofide's algorithms for TSP.

Social Network AnalysisGiorgos Cheliotis

This document provides an overview of social network analysis (SNA) including concepts, methods, and applications. It begins with background on how SNA originated from social science and network analysis/graph theory. Key concepts discussed include representing social networks as graphs, identifying strong and weak ties, central nodes, and network cohesion. Practical applications of SNA are also outlined, such as in business, law enforcement, and social media sites. The document concludes by recommending when and why to use SNA.

Self-Balancing Multimemetic Algorithms in Dynamic Scale-Free NetworksRafael Nogueras

A Performance Analysis of Self-* Evolutionary Algorithms on Networks with Cor...Rafael Nogueras

W+charm posterGiacomo Snidero

This document summarizes the measurement of W boson production in association with a charm quark using 4.6 pb-1 of proton-proton collision data at √s = 7 TeV from the ATLAS detector at the LHC. Two analyses are performed to tag charm quarks: one identifies charm jets through semileptonic decays to muons, and one reconstructs charmed mesons. Cross sections are reported and compared to theoretical predictions from different parton distribution functions. The results support PDF sets with an enhanced strange quark contribution and SU(3) symmetric light quark seas, and disfavor those with suppressed strange quarks. The measured cross sections have total uncertainties of 5-7% and indicate symmetric strange and anti-

Self-sampling Strategies for Multimemetic Algorithms in Unstable Computationa...Rafael Nogueras

This document discusses self-sampling strategies for multimemetic algorithms (MMAs) in unstable computational environments subject to churn. It proposes using probabilistic models to sample new individuals when populations need to be enlarged due to node failures. Experimental results show the bivariate model is superior for high churn, maintaining diversity and convergence better than random strategies. Future work aims to extend these self-sampling strategies to dynamic network topologies and more complex probabilistic models.

Evaluating the Synchronization of a Chaotic Encryption Scheme Using Different...IOSR Journals

This document evaluates the synchronization of a chaotic encryption scheme using different channel parameters through simulation. It simulates a chaotic encryption system based on Chua's chaotic oscillator circuits using Multisim software. The paper investigates the robustness of synchronization between a master-slave system by varying the resistance of the connecting line. The results show that synchronization can be achieved when the line resistance is below 3.5kΩ, but not above that value, limiting the potential distance between transmitter and receiver. Maintaining synchronization over channels is important for decrypting encrypted signals at the receiver.

F010123235IOSR Journals

This document evaluates the synchronization of a chaotic encryption scheme using different channel parameters through simulation. It summarizes that: 1) A master-slave synchronization scheme was achieved by connecting the capacitors of two identical Chua's chaotic circuits. 2) Perfect synchronization was achieved when the resistance of the connecting line was below 3.5kΩ. Higher resistances led to only partial or no synchronization. 3) Increasing the resistance of the connecting line beyond 3.5kΩ reduced the ability to synchronize the systems, limiting the distance between the transmitter and receiver for a chaotic encryption scheme. Amplification may help overcome higher resistances.

Sensitivity Analysis of Checkpointing Strategies for Multimemetic Algorithms ...Rafael Nogueras

FAULT DETECTION AND CLASSIFICATION ON SINGLE CIRCUIT TRANSMISSION LINE USING ...Politeknik Negeri Ujung Pandang

- This paper proposes a new technique of using discrete wavelet transform (DWT) and back-propagation neural network (BPNN) based on Clarke’s transformation for fault classification and detection on a single circuit transmission line. Simulation and training process for the neural network are done by using PSCAD / EMTDC and MATLAB. Daubechies4 mother wavelet (DB4) is used to decompose the high frequency components of these signals. The wavelet transform coefficients (WTC) and wavelet energy coefficients (WEC) for classification fault and detect patterns used as input for neural network training back-propagation (BPNN). This information is then fed into a neural network to classify the fault condition. A DWT with quasi optimal performance for preprocessing stage are presented. This study also includes a comparison of the results of training BPPN and DWT with and without Clarke’s transformation, where the results show that using Clarke transformation in training will give in a smaller mean square error (MSE) and mean absolute error (MAE). The simulation also shows that the new algorithm is more reliable and accurate.

Input analysisBhavik A Shah

The document discusses various methods for modeling input distributions in simulation models, including trace-driven simulation, empirical distributions, and fitting theoretical distributions to real data. It provides examples of several continuous and discrete probability distributions commonly used in simulation, including the exponential, normal, gamma, Weibull, binomial, and Poisson distributions. Key parameters and properties of each distribution are defined. Methods for selecting an appropriate input distribution based on summary statistics of real data are also presented.

ieee nss mic 2016 poster N30-21Dae Woon Kim

FAST実験3：新型大気蛍光望遠鏡の試験観測報告Toshihiro FUJII

The document describes the Fluorescence detector Array of Single-pixel Telescopes (FAST) project. FAST aims to build an array of low-cost single-pixel fluorescence detectors spaced over large areas to study ultra-high energy cosmic rays (UHECR). Each FAST station would have 12 telescopes with 4 PMTs each, covering a 30°x360° field of view. An array of 500 stations over 150,000 km2 could achieve an exposure over 12 times that of the Pierre Auger Observatory. Simulations show FAST may achieve 10% energy resolution and 35 g/cm2 Xmax resolution for cosmic rays above 1019.5 eV. The full-scale FAST prototype has been constructed and work is ongoing to develop

Recreation mathematics pptPawan Yadav

This document discusses various applications of graphs and networks including social networks, infrastructure networks, matching problems in bipartite graphs, scheduling conflicts, frequency assignment, register allocation, rectilinear pattern recognition, netlist layout, biomolecular constructions, DNA sequencing, and solving a "crazy cubes" puzzle. It provides examples and explanations of how each application can be modeled and solved as a graph problem. Key algorithms and techniques discussed include maximal matchings, graph coloring, subgraph search, network routing, and characterizing solutions to combinatorial problems.

NONLINEAR MODELING AND ANALYSIS OF WSN NODE LOCALIZATION METHODijwmn

In this paper, node localization algorithms in wireless sensor networks are researched, the traditional algorithms are studied, and some meaningful results are obtained. For the localization algorithm and route planning of WSN exists a big localization error in wireless communication. WSN communication system is researched. According to the anchor nodes and unknown nodes, a new localization algorithm and route planning method of WSN are proposed in this paper. At the same time, a new genetic algorithm of route planning of WSN is proposed. The performance of the node density and localization error is simulated and analyzed. The simulation results show that the performance of proposed WSN localization algorithm and route planning method are better than the traditional algorithms.

NONLINEAR MODELING AND ANALYSIS OF WSN NODE LOCALIZATION METHODijwmn

Cobb-douglas production functionSuniya Sheikh

The document describes the Cobb-Douglas production function and the results of estimating its parameters. It finds that labor (L) and capital (K) explain 95% of the variation in output (Q) according to the estimated equation Q=-0.31+0.20K+0.95L. Diagnostic tests show the data and estimated model meet the assumptions of regression analysis. Specifically, the variables are stationary and there is no multicollinearity while the model and estimated coefficients are statistically significant with a goodness of fit of 95%. Therefore, the Cobb-Douglas production function appropriately captures the relationship between L, K, and Q in the data.

Cobb-douglas production functionSuniya Sheikh

The document describes the Cobb-Douglas production function and the results of estimating its parameters. It finds that labor (L) and capital (K) explain 95% of the variation in output (Q) according to the estimated equation Q=-0.31+0.20K+0.95L. Diagnostic tests show the variables are stationary and there is no multicollinearity. The model is found to be statistically significant and a good fit to the data based on the F-statistic and R-squared value. It is determined that industry exhibits increasing returns to scale since the estimated coefficients of K and L sum to above 1.

Network switchingPREMAL GAJJAR

Du3211861191IJMER

This document reviews methods for enhancing the linearity of different types of transducers using artificial neural network (ANN) models. It identifies several ANN models that can be used for nonlinearity correction in resistive, capacitive, and inductive transducers, including functional link artificial neural network (FLANN), radial basis function neural network, multilayer perceptron, and backpropagation network. It finds that FLANN has advantages over these other models as it has lower computational complexity and is easier to implement. The document then provides specific examples of using backpropagation networks, FLANN, and multilayer perceptron models to improve the linearity of thermocouples, linear variable differential transformers, and capacitive pressure sensors, respectively.

08-Switching.pptSanaMateen7

This document discusses different types of switched networks including circuit-switched networks, datagram networks, and virtual-circuit networks. It provides details on the key aspects of each type: - Circuit-switched networks use dedicated channels for each connection and resources are allocated for the full duration of a call. Datagram networks divide messages into packets that are routed independently without resource reservation. Virtual-circuit networks have characteristics of both, allocating resources for the duration of a connection like circuit-switching but allowing packets to take different paths. - The structure of switches is also discussed, including crossbar switches for circuit switching and packet switches that use techniques like time-slot interchange and time-space-time switching to

Molecular Weight EstimationPhan Nghia

This document describes a method for determining the molecular weight of an unknown protein using SDS-PAGE (sodium dodecyl sulfate polyacrylamide gel electrophoresis). The method involves running protein standards and samples on an SDS-PAGE gel, measuring the migration distances, calculating Rf values, plotting a standard curve of log(MW) versus Rf, and using the equation generated to determine the MW of the unknown protein based on its Rf value. Key steps include separating proteins by SDS-PAGE, measuring band migration distances, calculating Rf, generating a standard curve, and using the curve equation to calculate the unknown MW.

B1102030610IOSR Journals

This document summarizes techniques for locating faults in underground power cables. It discusses using a microcontroller to determine the distance of a fault from the base station in kilometers by applying a low DC voltage through resistors representing the cable and detecting voltage drops. Tracer and terminal methods for fault location are described. Using fiber optic distributed temperature sensors and neural networks for fault identification are also summarized. The document provides block diagrams of the proposed underground cable fault distance system conveyed over GSM and its components including the power supply, microcontroller, rectifier, voltage regulator, and relay.

Path loss predictionNguyen Minh Thu

This summary provides an overview of the key points from the document: 1) The document presents the use of General Regression Neural Networks (GRNN) to predict propagation path loss in an urban environment based on measurements taken in Kavala, Greece. 2) Two neural network models are studied - one for path loss prediction and another using error control. Their performance is compared to measured path loss values based on error metrics. 3) For line-of-sight predictions, the GRNN model achieves better performance than empirical models due to using multiple input parameters and generalization. For non-line-of-sight, a third GRNN model including street orientation has the lowest error rates.

Computing on Event-sourced GraphsGraph-TA

This document discusses computing on event-sourced graphs. It presents Benjamin Erb from Ulm University who is researching combining event processing with graph computing. The document describes event sourcing and how it could be applied to graph data to maintain the history of a graph through event logs. This would allow reconstructing past states of the graph and running batch processes on different versions. Maintaining an event log could also power real-time and batch analytics on the same graph data.

Using Evolutionary Computing for Feature-driven Graph generationGraph-TA

1. The document discusses using evolutionary computing techniques to generate graphs with specific structural properties. 2. It presents an approach that uses genetic algorithms and neural networks to evolve graph generating functions that can produce graphs with user-specified properties like number of nodes, edges, clustering coefficient, etc. 3. The results show this approach can generate graphs faster than other evolutionary algorithms for graph generation, though it has trouble scaling to graphs with over 1,000 nodes.

More Related Content

Similar to Analysing the degree distribution of real graphs by means of several probabilistic models (20)

W+charm posterGiacomo Snidero

Self-sampling Strategies for Multimemetic Algorithms in Unstable Computationa...Rafael Nogueras

Evaluating the Synchronization of a Chaotic Encryption Scheme Using Different...IOSR Journals

F010123235IOSR Journals

Sensitivity Analysis of Checkpointing Strategies for Multimemetic Algorithms ...Rafael Nogueras

FAULT DETECTION AND CLASSIFICATION ON SINGLE CIRCUIT TRANSMISSION LINE USING ...Politeknik Negeri Ujung Pandang

Input analysisBhavik A Shah

ieee nss mic 2016 poster N30-21Dae Woon Kim

FAST実験3：新型大気蛍光望遠鏡の試験観測報告Toshihiro FUJII

Recreation mathematics pptPawan Yadav

NONLINEAR MODELING AND ANALYSIS OF WSN NODE LOCALIZATION METHODijwmn

Cobb-douglas production functionSuniya Sheikh

The document describes the Cobb-Douglas production function and the results of estimating its parameters. It finds that labor (L) and capital (K) explain 95% of the variation in output (Q) according to the estimated equation Q=-0.31+0.20K+0.95L. Diagnostic tests show the variables are stationary and there is no multicollinearity. The model is found to be statistically significant and a good fit to the data based on the F-statistic and R-squared value. It is determined that industry exhibits increasing returns to scale since the estimated coefficients of K and L sum to above 1.

Network switchingPREMAL GAJJAR

Du3211861191IJMER

08-Switching.pptSanaMateen7

Molecular Weight EstimationPhan Nghia

B1102030610IOSR Journals

Path loss predictionNguyen Minh Thu

W+charm posterGiacomo Snidero

Self-sampling Strategies for Multimemetic Algorithms in Unstable Computationa...Rafael Nogueras

Evaluating the Synchronization of a Chaotic Encryption Scheme Using Different...IOSR Journals

F010123235IOSR Journals

Sensitivity Analysis of Checkpointing Strategies for Multimemetic Algorithms ...Rafael Nogueras

FAULT DETECTION AND CLASSIFICATION ON SINGLE CIRCUIT TRANSMISSION LINE USING ...Politeknik Negeri Ujung Pandang

Input analysisBhavik A Shah

ieee nss mic 2016 poster N30-21Dae Woon Kim

FAST実験3：新型大気蛍光望遠鏡の試験観測報告Toshihiro FUJII

Recreation mathematics pptPawan Yadav

NONLINEAR MODELING AND ANALYSIS OF WSN NODE LOCALIZATION METHODijwmn

Cobb-douglas production functionSuniya Sheikh

Network switchingPREMAL GAJJAR

Du3211861191IJMER

08-Switching.pptSanaMateen7

Molecular Weight EstimationPhan Nghia

B1102030610IOSR Journals

Path loss predictionNguyen Minh Thu

More from Graph-TA (20)

Computing on Event-sourced GraphsGraph-TA

Using Evolutionary Computing for Feature-driven Graph generationGraph-TA

Reactive Databases for Big Data applicationsGraph-TA

This document discusses reactive databases for big data applications. It proposes an architecture with front-end and back-end components to support reactive programming and queries over NoSQL databases. The front-end would use a reactive language to declaratively define queries that observe changes. The back-end would integrate a rules engine with the NoSQL store to evaluate queries incrementally based on data changes and trigger notifications. Future work is outlined to optimize queries, integrate the system with Apache Spark, and implement more reactive language constructs.

The scarcity of crossing dependencies: a direct outcome of a specific constra...Graph-TA

This document summarizes a study on the scarcity of crossing dependencies in syntactic structures across languages. It presents two major hypotheses for why crossings are scarce: 1) an underlying rule or principle prohibits crossings, or 2) crossings are indirectly limited by dependency length minimization, which constrains dependency lengths. The study evaluates these hypotheses by analyzing dependency structures from 30 languages and finding that accounting for dependency lengths reduces errors in predicting crossings compared to random arrangements, supporting the second hypothesis.

Holistic Benchmarking of Big Linked Data: HOBBITGraph-TA

The document discusses the HOBBIT project, which aims to provide holistic benchmarking of big linked data technologies. It notes the large volumes of linked data and number of available technologies. HOBBIT will identify key performance indicators and benchmarks for critical steps in the linked data lifecycle, including generation, analysis, storage, and visualization. It will gather industry performance needs and provide a platform for standardized benchmarking and challenges to compare technology performance. The project is coordinated by researchers in Europe and runs for three years under the Horizon 2020 program.

Identifiability in Dynamic Casual NetworksGraph-TA

The document discusses algorithms for identifying causal effects in dynamic causal networks (DCNs). It presents the DCN-ID algorithm, which can identify causal effects in DCNs with either static or dynamic confounders. For DCNs with static confounders, the algorithm shows that the network recovers time-invariant behavior after an intervention. However, for DCNs with dynamic confounders, the algorithm reveals that the network does not necessarily recover time-invariant behavior after intervention. The algorithm provides a complete method for identifying causal effects in DCNs regardless of whether confounders are static or dynamic.

Polyglot Graph Databases using OCL as pivotGraph-TA

This document discusses using the Object Constraint Language (OCL) as a pivot language for polyglot graph databases. OCL could serve as a common query language that is translatable to different native graph languages. This would eliminate the need for code generation between languages and allow a graph database to support multiple query languages. The document also proposes extensions to OCL, such as adding syntax for graph patterns, to make it more suitable as a graph query language.

Benchmarking Versioning for Big Linked DataGraph-TA

The document discusses benchmarking versioning systems for big linked data. It describes the need for versioning benchmarks that test how systems perform with respect to storage space required for multiversion repositories and efficiency of retrieving different versions and answering cross-snapshot queries. It outlines approaches for versioning like full materialization, delta-based, and timestamped tuples. The document also lists several linked data stores with versioning capabilities and notes the complete lack of existing versioning benchmarks. It proposes the design of a versioning benchmark for the HOBBIT project that would generate versions based on changes observed in evolving datasets and include cross-snapshot queries to evaluate system performance.

Synthetic Data Generation using exponential random Graph modelingGraph-TA

This document discusses using exponential random graph modeling to generate synthetic network data. It introduces exponential random graph models and how they use log-linear models and conditional log-odds to model network formation. An example network of tobacco control agencies is analyzed, including adding node attributes, edge attributes, and structural effects to the model. Model diagnostics and goodness of fit are evaluated. Finally, the trained model can be used to simulate new synthetic networks.

Use of Graphs for Cloud Service Selection in Multi-Cloud EnvironmentsGraph-TA

This document discusses using graphs to model relationships between cloud services from different providers to help with service selection for multi-cloud environments. It presents graphs as a way to capture compatibility, replaceability, and other conditional relationships between services. Centrality metrics, community detection algorithms, and graph pattern matching could then be applied to the modeled graphs to analyze service configurations and identify options that meet requirements for adaptability, quality, and cost. An example graph instantiation is provided to demonstrate compatibility modeling between specific database and application services from Amazon, Google, Microsoft, and other cloud providers.

Graphalytics: A big data benchmark for graph-processing platformsGraph-TA

Graphalytics is a benchmark for evaluating graph processing platforms. It includes a diverse set of algorithms and synthetic and real-world datasets. The benchmark harness collects performance metrics across platforms and enables in-depth bottleneck analysis through Granula. Graphalytics aims to enable fair comparison of different graph systems and help identify areas for improvement through a modern software development process.

Modelling the Clustering Coefficient of a Random graphGraph-TA

1) The document describes an algorithm for generating random graphs that have a specified degree distribution and average clustering coefficient. 2) The algorithm works by splitting the graph into communities and connecting nodes within and between communities with different probabilities to achieve the target average clustering coefficient. 3) An extended hypergeometric distribution is used to model the probability of connections for the highest degree node, and the expected clustering coefficient is set equal to the target value to determine the connection probabilities.

RDF Graph Data Management in Oracle Database and NoSQL PlatformsGraph-TA

GRAPHITE — An Extensible Graph Traversal Framework for RDBMSGraph-TA

This document discusses GRAPHITE, an extensible graph traversal framework for relational database management systems (RDBMS). GRAPHITE allows for efficient graph processing by integrating native graph operators into the RDBMS query execution engine. This allows graph queries to be combined with relational and other data models within a single database. The architecture of GRAPHITE includes components for graph storage, logical and physical graph operators, a graph optimizer, and a graph traversal DSL called TraveL. Initial experimental results show that GRAPHITE can efficiently execute graph traversals and queries on graph data stored in an RDBMS.

On the Discovery of Novel Drug-Target Interactions from Dense SubGraphsGraph-TA

This document discusses the discovery of novel drug-target interactions from dense subgraphs in bipartite graphs representing drug-target networks. It presents the empirical subgraph density (esDSG) problem, which aims to identify the maximally dense subgraph where novel interactions could be predicted. The authors evaluate their approach on a drug-target network dataset containing over 5,000 known interactions across four classes of targets. Their results demonstrate the esDSG problem can suggest interactions that other methods cannot. Future work will involve extending esDSG to identify dense subgraphs across different connected components.

Graphalytics: A big data benchmark for graph processing platformsGraph-TA

Graphalytics is an advanced benchmark harness for evaluating graph processing platforms. It performs choke-point analysis to identify key technological challenges. The benchmark workload targets these choke points to assess performance. It also features an enhanced LDBC Datagen tool that generates graphs with varying structural characteristics. Graphalytics aims to provide a standardized way to benchmark and compare different graph processing platforms.

Autograph: an evolving lightweight graph toolGraph-TA

The document discusses the motivation for developing a new graph tool called Autograph. It notes that existing graph software libraries and tools have limitations and frustrations for users who want an easy way to draw graphs and obtain basic graph properties. The document outlines a wish list of desired features for Autograph, including easy graph editing, undo/redo, geometric transformations, real-time updating of properties, and basic graph operations. It includes screenshots and poses questions to further develop Autograph.

Understanding Graph Structure in Knowledge BasesGraph-TA

This document discusses analyzing the graph structure of Wikipedia to improve query expansion techniques. It describes using Wikipedia as a knowledge base with millions of articles and extracting semantic information from queries by analyzing related articles, categories, and redirects in the Wikipedia graph. The goal is to understand relationships between different data types in Wikipedia and how this can help identify semantically related information to expand queries and improve search results.

Finding patterns of chronic disease and medication prescriptions from a large...Graph-TA

Recent Updates on IBM System G — GraphBIG and Temporal DataGraph-TA

This document discusses recent updates on IBM System G and temporal graph data. It introduces GraphBIG, a graph benchmark suite, and how it can be used to evaluate the performance of different processor architectures for graph workloads. It also discusses challenges in managing temporal graph data that evolves over time and supporting bitemporal data exploration. Methods for organizing temporal graph data storage and performing parallel temporal graph inference using pointer jumping for Gaussian elimination are presented.