Graph Data Mining Research Articles

MvInf: Social Influence Prediction with Multi-view Graph Attention Learning

The potential impact of social influence prediction has become a hot topic in the current graph data mining area. This paper proposes a deep learning framework named Multi-view Influence prediction network (MvInf) which combines multi-view learning and graph attention neural network together to address the problem of social influence prediction. MvInf takes different attribute features of users as the input of graph attention network and uses the complementarity and consistency between different views to enhance learning performance and thus to better predict user behavior. Experiments performed on four standard datasets (Open Academic Graph, Twitter, Weibo, and Digg) demonstrate that the proposed MvInf model can obtain better performance than previous single view-based approach.

Automated Unsupervised Graph Representation Learning

Graph data mining has largely benefited from the recent developments of graph representation learning. Most attempts to improve graph representations have thus far focused on designing new network embedding or graph neural network (GNN) architectures. Inspired by the SGC and ProNE models, we instead focus on enhancing any existing or learned graph representations by further smoothing them via graph filters. In this paper, we introduce an automated framework AutoProNE to achieve this. Specifically, AutoProNE automatically searches for a unique optimal set of graph filters for any input dataset, and its existing representations are then smoothed via the selected filters. To make AutoProNE more general, we adopt self-supervised loss functions to guide the optimization of the automated search process. Extensive experiments on eight commonly used datasets demonstrate that the AutoProNE framework can consistently improve the expressive power of graph representations learned by existing network embedding and GNN methods by up to 44%.

Mining user–user communities for a weighted bipartite network using spark GraphFrames and Flink Gelly

Large-scale graph processing is one of the recently developed significant research areas relevant to big data analytics. Distributed graph analytics is useful to see the intuitive insights of node interactions from large-scale network data. Distributed graph computing is an upcoming area in graph data mining that explores crucial node relationships for a given graph dataset. In this paper, we propose a new method to discover top-k user–user communities for a weighted bipartite network by defining a weighted similarity measure. We extend the structural similarity metric, namely Otsuka–Ochiai coefficient, by adding weights of nodes and quantifies the similarity between distinct items of a user–item network. We propose a new method to mine top-k user–user communities based on the similarity of items using a weighted similarity measure. Further, two algorithms, namely TUCSGF, TUCFlink, are presented to mine top-k user–user communities in a distributed approach based on the strength of the item-to-item similarities. Moreover, we execute the TUCSGF algorithm using Apache Spark by utilizing the advantage of Spark GraphFrames to mine top-k user–user communities. Also, we implement the TUCFlink algorithm to mine top-k communities using Apache Flink by utilizing the functionalities of Flink Gelly. Further, we explore two real-world network applications online learning network, chain of hospitals network with various graph methods that are to be applied for both the applications. Furthermore, we systematically perform various experiments concerning execution time, memory consumption, and CPU usage of both TUCSGF, TUCFlink on three distinct datasets. The performance of TUCFLINK is far better than TUCSGF concerning computing time. Applying distributed graph analytics for various complex networks using distributed graph processing tools GraphX, GraphFrames and Gelly provides more intuitive insights about distinct types of node interactions in graph data mining.

Analysis of Internal and External Academic Collaboration in an Institution Through Graph Theory

This paper presents an analysis of scientific collaboration through graph theory, based on a previous study focused on the collaborative work of researchers within an institution. This proposal also exposes the representation of inter-institutional collaboration of research groups, combining graph theory and data mining. The state of the art relates the concepts of scientific production, digital repositories, interoperability between repositories, the law of Open Science in Mexico, the theory of graphs and their use in previous studies for the analysis of scientific collaboration, and the definition of research groups in Mexico. Furthermore, the methodology uses elements of knowledge extraction for data mining, involving recovery, processing and visualization. Results present the collaboration status at the Universidad Autónoma de Yucatán, internally and externally, by the research groups. Internally, 22 groups were found and each researcher collaborates with six other professors within the institution, on average. In addition, consolidated research groups are those with the highest level of production and collaboration at national and international level, compared to the groups with less consolidation.

RETRACTED ARTICLE: Speech big data cluster modeling based on joint neural network and Spark-SVM with evolutionary intelligence

Emerging development of artificial intelligence application scenarios have caused the bursting requirement for the speech analysis tools, the efficient speech data processing model is urgently needed. This paper analyzes the speech big data cluster modeling based on joint neural network and Spark-SVM. As one of the important functions of graph data mining and analysis applications, graph clustering mainly implements classification operations on each node in the graph model based on clusters, and at the same time increases the association of the object entities, we then use this feature to design the model. In practical applications, subject to the size of the training data set, the system recognition rate does not show a steady upward trend with the increase of Gaussian mixture. In the case of limited speech data, the model parameters that can be then reliably estimated are limited. Because the input vector in the neural network structure is mostly abstract data, the BN layer in the hidden layer must be located in the back of the network structure, which makes the hierarchical performance results more profound. Hence, the joint neural network model is designed. The Spark structure is implemented to improve the systematic efficiency. We simulate the model and compare with the state-of-the-art models.

Online social network trend discovery using frequent subgraph mining

Graph mining has become a well-established discipline within the domain of data mining. It has received much interest over the last decade as advances in computer hardware have provided the processing power to enable large-scale graph data mining to be conducted. Frequent subgraph mining (FSM) plays a very significant role in graph mining, attracting a great deal of attention in different domains, such as Bioinformatics, web data mining and social networks. Online social networks (SNs) play an important role in today’s Internet. These social networks contain huge amounts of data and present a challenging problem. FSM has been used in SNs to identify the frequent pattern trends existing in the network. A frequent pattern trend is defined as a sequence of time-stamped occurrences (support) value for specific frequent pattern that exist in the data. For example, most active researchers, most visited web pages or users’ navigation patterns over the web are few to mention. In the past few years, social network trend mining has been an active area of research. Many graph mining algorithms have been proposed, but a very limited effort exists for capturing an important dimension of SNs, which is trends discovery. Therefore, this paper introduces a novel FSM approach, called A-RAFF (A RAnked Frequent pattern-growth Framework), to discovering and comparing the frequent pattern trends exist in the social network data. Furthermore, the social network frequent pattern trend analysis has been evaluated using two standard social networks, Facebook-like network and the famous MSNBC news network datasets. Consequently, the discovered trends will help the underlying social networks to further enhance their platforms for the betterment of the users as well as for their business growth.

Faster Algorithms for Mining Shortest-Path Distances from Massive Time-Evolving Graphs

Computing shortest-path distances is a fundamental primitive in the context of graph data mining, since this kind of information is essential in a broad range of prominent applications, which include social network analysis, data routing, web search optimization, database design and route planning. Standard algorithms for shortest paths (e.g., Dijkstra’s) do not scale well with the graph size, as they take more than a second or huge memory overheads to answer a single query on the distance for large-scale graph datasets. Hence, they are not suited to mine distances from big graphs, which are becoming the norm in most modern application contexts. Therefore, to achieve faster query answering, smarter and more scalable methods have been designed, the most effective of them based on precomputing and querying a compact representation of the transitive closure of the input graph, called the 2-hop-cover labeling. To use such approaches in realistic time-evolving scenarios, when the managed graph undergoes topological modifications over time, specific dynamic algorithms, carefully updating the labeling as the graph evolves, have been introduced. In fact, recomputing from scratch the 2-hop-cover structure every time the graph changes is not an option, as it induces unsustainable time overheads. While the state-of-the-art dynamic algorithm to update a 2-hop-cover labeling against incremental modifications (insertions of arcs/vertices, arc weights decreases) offers very fast update times, the only known solution for decremental modifications (deletions of arcs/vertices, arc weights increases) is still far from being considered practical, as it requires up to tens of seconds of processing per update in several prominent classes of real-world inputs, as experimentation shows. In this paper, we introduce a new dynamic algorithm to update 2-hop-cover labelings against decremental changes. We prove its correctness, formally analyze its worst-case performance, and assess its effectiveness through an experimental evaluation employing both real-world and synthetic inputs. Our results show that it improves, by up to several orders of magnitude, upon average update times of the only existing decremental algorithm, thus representing a step forward towards real-time distance mining in general, massive time-evolving graphs.

Design of affinity-aware encoding by embedding graph centrality for graph classification

Deep learning methods for graph classification are critical for graph data mining. Recently, graph convolutional networks (GCNs) have been able to achieve state-of-the-art node classification. A typical process for GCNs includes two iterative steps: node feature encoding and message passing. While the former encodes each graph node independently via the uniform encoding function, the latter updates the features of each node by weighted aggregation of the features of neighboring nodes, where the weights are generated by predefined or learned graph Laplacian. However, their accuracy deteriorates for graph classification tasks because the uniform encoding function encodes all the node features involved. In this study, we propose a novel affinity-aware encoding for graph classification. In our model, we implement a separate encoding function for the neighboring nodes of each node for updating the node features, where the nodes are arranged in the order of affinity values, such as graph centrality, in order to determine the correspondence between an encoding function and a specific neighboring node. Our separate encoding function performs non-Euclidean neighboring encoding for each node by weight sharing, which enables message passing. We also develop two variants based on our separate encoding function: the graph centrality-convolutional neural network (C-CNN) and the graph centrality-graph convolutional network (C-GCN). The former performs the separate encoding function on graph data directly by the function of message passing. The latter combines the separate encoding function with the normalized graph Laplacian implemented on the graph data. Experiments demonstrate that the results obtained by our models are consistent with those obtained by classical convolutional neural networks (CNNs) on the MNIST dataset, and they outperform existing GCNs on the 20NEWS, Reuters8, and Reuters52 datasets. We also apply our two variants to online car-hailing service data for traffic congestion recognition. Our methods show state-of-the-art results compared with GCNs.

An Empirical Study of Deep Web based on Graph Analysis

The internet can broadly be divided into three parts: surface, deep and dark among which the latter offers anonymity to its users and hosts [1]. Deep Web refers to an encrypted network that is not detected on search engine like Google etc. Users must use Tor to visit sites on the dark web [2]. Ninety six percent of the web is considered as deep web because it is hidden. It is like an iceberg, in that, people can just see a small portion above the surface, while the largest part is hidden under the sea [3, 4, and 5]. Basic methods of graph theory and data mining, that deals with social networks analysis can be comprehensively used to understand and learn Deep Web and detect cyber threats [6]. Since the internet is rapidly evolving and it is nearly impossible to censor the deep web, there is a need to develop standard mechanism and tools to monitor it. In this proposed study, our focus will be to develop standard research mechanism to understand the Deep Web which will support the researchers, academicians and law enforcement agencies to strengthen the social stability and ensure peace locally & globally.

High performance frequent subgraph mining on transaction datasets: A survey and performance comparison

Graph data mining has been a crucial as well as inevitable area of research. Large amounts of graph data are produced in many areas, such as Bioinformatics, Cheminformatics, Social Networks, etc. Scalable graph data mining methods are getting increasingly popular and necessary due to increased graph complexities. Frequent subgraph mining is one such area where the task is to find overly recurring patterns/subgraphs. To tackle this problem, many main memory-based methods were proposed, which proved to be inefficient as the data size grew exponentially over time. In the past few years, several research groups have attempted to handle the Frequent Subgraph Mining (FSM) problem in multiple ways. Many authors have tried to achieve better performance using Graphic Processing Units (GPUs) which has multi-fold improvement over in-memory while dealing with large datasets. Later, Google's MapReduce model with the Hadoop framework proved to be a major breakthrough in high performance large batch processing. Although MapReduce came with many benefits, its disk I/O and noniterative style model could not help much for FSM domain since subgraph mining process is an iterative approach. In recent years, Spark has emerged to be the De Facto industry standard with its distributed in-memory computing capability. This is a right fit solution for iterative style of programming as well. In this survey, we cover how high-performance computing has helped in improving the performance tremendously in the transactional directed and undirected aspect of graphs and performance comparisons of various FSM techniques are done based on experimental results.

Deep mining port scans from darknet

SummaryTCP/UDP port scanning or sweeping is one of the most common technique used by attackers to discover accessible and potentially vulnerable hosts and applications. Although extracting and distinguishing different port scanning strategies is a challenging task, the identification of dependencies among probed ports is primordial for profiling attacker behaviors, with a final goal of better mitigating them. In this paper, we propose an approach that allows to track port scanning behavior patterns among multiple probed ports and identify intrinsic properties of observed group of ports. Our method is fully automated based on graph modeling and data mining techniques, including text mining.It provides to security analysts and operators relevant information about services that are jointly targeted by attackers. This is helpful to assess the strategy of the attacker by understanding the types of applications or environment he or she targets. We applied our method to data collected through a large Internet telescope (or darknet).

Link Prediction in Evolving Networks Base on Information Propagation

Link prediction is an important issue in graph data mining. In social networks, link prediction is used to predict missing links in current networks and new links in future networks. This process has a wide range of applications including recommender systems, spam mail classification, and the identification of domain experts in various research areas. In order to predict future node similarity, we propose a new model, Common Influence Set, to calculate node similarities. The proposed link prediction algorithm uses the common influence set of two unconnected nodes to calculate a similarity score between the two nodes. We used the area under the ROC curve (AUC) to evaluate the performance of our algorithm and that of previous link prediction algorithms based on similarity over a range of problems. Our experimental results show that our algorithm outperforms previous algorithms.

A novel parallel distance metric-based approach for diversified ranking on large graphs

Diversified ranking on graphs (DRG) is an important and challenging issue in researching graph data mining. Traditionally, this problem is modeled by a submodular optimization objective, and solved by applying a cardinality constrained monotone submodular maximization. However, the existing submodular objectives do not directly capture the dis-similarity over pairs of nodes, while most of algorithms cannot easily take full advantage of the power of a distributed cluster computing platform, such as Spark, to significantly promote the efficiency of algorithms. To overcome the deficiencies of existing approaches, in this paper, a generalized distance metric based on a subadditive set function over the symmetry difference of neighbors of pairs of nodes is introduced to capture the pairwise dis-similarity over pairs of nodes. In our approach,DRG is formulated as a Max-Sum k-dispersion problem with metrical edge weights, which is NP-hard, in association with the proposed distance metric, a centralized linear time 2-approximation algorithm GA is then developed to significantly solve the problem ofDRG. Moreover, we develop a highly parallelizable algorithm forDRG, which can be easily implemented in MapReduce style parallel computation models using GA as a basic reducer. Finally, extensive experiments are conducted on real network datasets to verify the effectiveness and efficiency of our proposed approaches.

A task‐based approach for finding SCCs in real‐world graphs on external memory

SummaryFinding strongly connected components (SCCs) in graphs is one of the important research topics of graph data mining. Traditional SCC‐finding methods need to load the whole graph into main memory before actual processing, which makes them inappropriate to process today's large graphs. Although it is cost‐effective to conductive SCC‐finding in the external memory (EM) graph processing systems, existing EM systems are still inefficient on such workload for 2 reasons: data‐parallel processing model and inefficient graph mutation mechanism.In this paper, we propose a task‐based approach, named as TAS, for finding SCCs in large real‐world graphs. TAS encapsulates individual graph dataset and the SCC‐finding algorithms conducted on it as an independent task. By this strategy, different algorithms can be assigned to different datasets according to the stage of processing or the size of the dataset. By gradually removing unneeded data, ie, the known SCCs, with an efficient graph mutation method, TAS continuously reduces the scale of the problem to accelerate processing. Performance evaluation on large real‐world graphs show that TAS greatly outperforms existing EM solutions on finding SCCs (eg, 55.2x faster than GraphChi and 40.3x faster than X‐Stream).

Detecting cyber threats through social network analysis: short survey

This article considers a short survey of basic methods of social networks analysis, which are used for detecting cyber threats. The main types of social network threats are presented. Basic methods of graph theory and data mining, that deals with social networks analysis are described. Typical security tasks of social network analysis, such as community detection in network, detection of leaders in communities, detection experts in networks, clustering text information and others are considered.

Para-G: Path pattern query processing on large graphs

There are plentiful and diverse applications of graph data management and mining techniques in the real-world scientific research and business activities. As one of the most basic operations, uniform path pattern query processing on graph data faces three big challenges. In this paper, we deal with these challenges by the following points. Firstly, a new query language on graph, called G-Path, is presented, which focuses on complex path pattern query processing on a very large graph. Also, the design of a system called Para-G is proposed, which is based on a BSP-like model as well as MapReduce model, and can effectively handle distributed graph data operations and queries. Secondly, the implementation of Para-G on the de facto cloud platform -- Hadoop -- is brought forward. Based on the concept of distributed path finite state automaton, the query processing of a G-Path statement in Para-G is detailed. In addition, as the query optimization of G-Path queries, several tricks are utilized to dramatically improve the performance of query execution. Finally, extensive experiments on several graph data sets are conducted to show the usability of the G-Path query language and the effectiveness of Para-G.

Put Three and Three Together

Community detection has arisen as one of the most relevant topics in the field of graph data mining due to its applications in many fields such as biology, social networks, or network traffic analysis. Although the existing metrics used to quantify the quality of a community work well in general, under some circumstances, they fail at correctly capturing such notion. The main reason is that these metrics consider the internal community edges as a set, but ignore how these actually connect the vertices of the community. We propose the Weighted Community Clustering ( WCC ), which is a new community metric that takes the triangle instead of the edge as the minimal structural motif indicating the presence of a strong relation in a graph. We theoretically analyse WCC in depth and formally prove, by means of a set of properties, that the maximization of WCC guarantees communities with cohesion and structure. In addition, we propose Scalable Community Detection (SCD) , a community detection algorithm based on WCC , which is designed to be fast and scalable on SMP machines, showing experimentally that WCC correctly captures the concept of community in social networks using real datasets. Finally, using ground-truth data, we show that SCD provides better quality than the best disjoint community detection algorithms of the state of the art while performing faster.

Expert system for crop disease based on graph pattern matching: a proposal

For agroindustry, crop diseases constitute one of the most common problems that generate large economic losses and low production quality. On the other hand, from computer science, several tools have emerged in order to improve the prevention and treatment of these diseases. In this sense, recent research proposes the development of expert systems to solve this problem, making use of data mining and artificial intelligence techniques like rule-based inference, decision trees, Bayesian network, among others. Furthermore, graphs can be used for storage of different types of variables that are present in an environment of crops, allowing the application of graph data mining techniques like graph pattern matching. Therefore, in this paper we present an overview of the above issues and a proposal of an expert system for crop disease based on graph pattern matching.

A Sub-graph Matching Method based on Calibration of Characteristics of Topological Footprint

Approximate sub-graph matching is important in many graph data mining fields. At present, current solutions can be difficult to implement, have an expensive pre-processing phase, or only work for given types of graph. In this paper a novel generic approach is presented which addresses these issues. An approximate sub-graph matcher (A-SGM) calculates the distance between the topological characteristics (footprint) of the sub-graphs to be matched, applying a weighting to the different sub-graph characteristics and those of neighbor nodes. The weights are calibrated for each dataset with a simulated annealing process using sample sets of graph nodes to reduce computational cost, and an exact isomorphism matcher as a fitness function which takes into account how well the match maintains the neighboring node degree distributions. Benchmarking is performed on several state of the art methods and real and synthetic graph datasets to evaluate the precision, recall and computational cost. The results show that the A-SGM is competitive with state of the art methods in terms of precision, recall and execution time. General Terms Machine Intelligence (Data and Web Mining), Applications of Computer Science in Modeling, Data and Information Systems.

Accuracy estimation of link-based similarity measures and its application

Link-based similarity measures play a significant role in many graph based applications. Consequently, measuring node similarity in a graph is a fundamental problem of graph datamining. Personalized pagerank (PPR) and simrank (SR) have emerged as the most popular and influential link-based similarity measures. Recently, a novel link-based similarity measure, penetrating rank (P-Rank), which enriches SR, was proposed. In practice, PPR, SR and P-Rank scores are calculated by iterative methods. As the number of iterations increases so does the overhead of the calculation. The ideal solution is that computing similarity within the minimum number of iterations is sufficient to guarantee a desired accuracy. However, the existing upper bounds are too coarse to be useful in general. Therefore, we focus on designing an accurate and tight upper bounds for PPR, SR, and P-Rank in the paper. Our upper bounds are designed based on the following intuition: the smaller the difference between the two consecutive iteration steps is, the smaller the difference between the theoretical and iterative similarity scores becomes. Furthermore, we demonstrate the effectiveness of our upper bounds in the scenario of top-k similar nodes queries, where our upper bounds helps accelerate the speed of the query. We also run a comprehensive set of experiments on real world data sets to verify the effectiveness and efficiency of our upper bounds.