ScaleSCAN: Scalable Density-Based Graph Clustering

Shiokawa, Hiroaki; Takahashi, Tomokatsu; Kitagawa, Hiroyuki

doi:10.1007/978-3-319-98809-2_2

Cited by 20 publications

(6 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…GOrder [41] (commit 7ccdfe9), Rabbit-Order [2] (commit f67a79e), and SlashBurn [24] are state-of-the-art locality optimizing relabeling algorithms we use for evaluation of iHTL. GOrder has a limit of |𝐸| < 2 31 and Rabbit-Order could not complete relabeling of ClWb9 because of an "out of memory" error.…”

Section: Evaluation Methods and Datasetsmentioning

confidence: 99%

“…Scan isolates hubs and outliers (vertices marginally appended to clusters) from clusters to prevent unrelated communities to be merged because of only one hub neighbour. ScaleScan [31] removes unnecessary computations in Scan and parallelizes the execution. Graph relabeling algorithms like SlashBurn [24], GOrder [41] and Rabbit-Order [2] rearrange vertices in order to improve locality.…”

Section: Locality Optimizing Graph Reorderingmentioning

confidence: 99%

See 1 more Smart Citation

Exploiting in-Hub Temporal Locality in SpMV-based Graph Processing

Esfahani

Kilpatrick

Vandierendonck

2021

50th International Conference on Parallel Processing

View full text Add to dashboard Cite

The skewed degree distribution of real-world graphs is the main source of poor locality in traversing all edges of the graph, known as Sparse Matrix-Vector (SpMV) Multiplication. Conventional graph traversal methods, such as push and pull, traverse all vertices in the same manner, and we show applying a uniform traversal direction for all edges leads to sub-optimal memory locality, hence poor efficiency. This paper argues that different vertices in power-law graphs have different locality characteristics and the traversal method should be adapted to these characteristics.To solve this problem, we propose to inspect the number of destination and source vertices in selecting a cachecompatible traversal direction for each type of vertex. We introduce in-Hub Temporal Locality (iHTL), a structure-aware SpMV that combines push and pull in one graph traversal, but for different vertex types. iHTL exploits temporal locality by traversing incoming edges to in-hubs in push direction, while processing other edges in pull direction.The evaluation shows iHTL is 1.5× -2.4× faster than pull and 4.8× -9.5× faster than push in state-of-the-art graph processing frameworks such as GraphGrind, GraphIt and Galois. More importantly, iHTL is 1.3× -1.5× faster than pull traversal of state-of-the-art locality optimizing reordering algorithms such as SlashBurn, GOrder, and Rabbit-Order.

show abstract

Section: Evaluation Methods and Datasetsmentioning

confidence: 99%

Section: Locality Optimizing Graph Reorderingmentioning

confidence: 99%

Exploiting in-Hub Temporal Locality in SpMV-based Graph Processing

Esfahani

Kilpatrick

Vandierendonck

2021

50th International Conference on Parallel Processing

View full text Add to dashboard Cite

show abstract

“…Although modularity clustering can efficiently handle large-scale graphs, it can not fully reproduce the ground-truth clusters [8]. To overcome the low accuracy of modularity clustering algorithms, density-based graph clustering algorithms [23], [27], [28] have been proposed in recent years. Density-based graph clustering is an extension of DBSCAN [7], which is a traditional density-based clustering algorithm for multi-dimensional data objects.…”

Section: Related Workmentioning

confidence: 99%

“…The problem of finding clusters in a graph has been studied for several decades in many areas, especially in physics and computer science. Traditional clustering algorithms such as graph partitioning [18], modularity clustering [19], [20], and densitybased methods [21], [23], [27] are natural choices for this problem. Basically, these algorithms are designed to compute homogeneous graphs.…”

Section: Introductionmentioning

confidence: 99%

Fast and Parallel RankClus Algorithm based on Dynamic Rank Score Tracking

Yamazaki

Matsugu

Shiokawa

et al. 2020

Journal of Information Processing

Self Cite

View full text Add to dashboard Cite

Given a bi-type information network, which is an extended model of well-known bipartite graphs, how can clusters be efficiently found in graphs? Graph clustering is now a fundamental tool to understand overviews from graph-structured data. The RankClus framework accurately performs clustering for bi-type information networks using ranking-based graph clustering techniques. It integrates graph ranking algorithms such as PageRank or HITS into graph clustering procedures to improve the clustering quality. However, this integration incurs a high computational cost to handle large bi-type information networks since RankClus repeatedly computes the ranking algorithm for all nodes and edges until the clustering procedure converges. To overcome this runtime limitation, we present a novel RankClus algorithm that reduces the running time for large bi-type information networks. Our proposed method employs dynamic graph processing techniques into the iterative ranking procedures included in RankClus. By dynamically updating ranking results, our proposal reduces the number of computed nodes and edges during repeated ranking procedures. For further improving the efficiency, we also present a parallel implementation of our proposed algorithm by using thread-based parallelization on a modern manycore CPU. We experimentally verify using real-world datasets that our fast and parallel proposed methods successfully reduces the running time while maintaining the clustering quality of RankClus.

show abstract

“…Once they receive a query node from a user, the community search algorithms explore a single community (cluster) that has dense inner-community connections with the largest relevance to the query node. Unlike traditional community detection algorithms (e.g., modularity-based methods [1], [14] and densitybased methods [16], [17]), community search algorithms can return a search result within a short computation time since they do not need to compute the entire graph. Due to their efficiency, such algorithms have been applied to various applications, including social analysis and protein analysis.…”

Section: Introductionmentioning

confidence: 99%

Fast Algorithm for Attributed Community Search

Matsugu

Shiokawa

Kitagawa

2021

Journal of Information Processing

Self Cite

View full text Add to dashboard Cite

Searching communities on attributed graphs has attracted much attention in recent years. The community search algorithm is currently an essential graph data management tool to find a community suited to a user-specified query node. Although community search algorithms are useful in various web-based applications and services, they have trouble handling attributed graphs due to the strict topological constraints of traditional algorithms. In this paper, we propose an accurate community search algorithm for attributed graphs. To relax the topological constraints, we proposed a new model of the community. And we defined the problem of finding them in an attributed graph class called the Flexible Attributed Truss Community (F-ATC). The F-ATC problem has the advantage of being applicable in many situations because it can explore diverse communities. Consequently, the community search accuracy is enhanced compared to traditional community search algorithms. Additionally, we present a novel heuristic algorithm to solve the F-ATC problem. This effective algorithm detects more accurate communities from attributed graphs than the traditional algorithms. For further optimization, we pre-processed the query response to make it faster. Finally, we conducted extensive experiments with real-world attributed graphs to demonstrate that our approach outperforms state-of-the-art methods.

show abstract

ScaleSCAN: Scalable Density-Based Graph Clustering

Cited by 20 publications

References 17 publications

Exploiting in-Hub Temporal Locality in SpMV-based Graph Processing

Exploiting in-Hub Temporal Locality in SpMV-based Graph Processing

Fast and Parallel RankClus Algorithm based on Dynamic Rank Score Tracking

Fast Algorithm for Attributed Community Search

Contact Info

Product

Resources

About