随机随行社区关键成员搜索大图 (Random Walk-based Community Key-members Search over Large Graphs)

from arxiv, In the current version, we only include an accuracy analysis, not a hard accuracy guarantee. This may confuse readers that we are sensitive to the dataset. So, we plan to re-design the key (i.e., transition matrix) of our random walk-based solution, thus providing an accuracy guarantee (e.g., upper/lower bound). We expect to withdraw this version and pay more time to re-consider this issue

Given a graph $G$, a query node $q$, and an integer $k$, community search (CS) seeks a cohesive subgraph (measured by community models such as $k$-core or $k$-truss) from $G$ that contains $q$. It is difficult for ordinary users with less knowledge of graphs' complexity to set an appropriate $k$. Even if we define quite a large $k$, the community size returned by CS is often too large for users to gain much insight about it. Compared against the entire community, key-members in the community appear more valuable than others. To contend with this, we focus on a new problem, that is \textbf{C}ommunity \textbf{K}ey-members \textbf{S}earch problem (CKS). We turn our perspective to the key-members in the community containing $q$ instead of the entire community. To solve CKS problem, we first propose an exact algorithm based on truss decomposition as the baseline. Then, we present four random walk-based optimized algorithms to achieve a trade-off between effectiveness and efficiency, by carefully considering some important cohesiveness features in the design of transition matrix. We return the top-$n$ key-members according to the stationary distribution when random walk converges. Moreover, we propose a lightweight refinement method following an "expand-replace" manner to further optimize the top-$n$ result with little overhead, and we extend our solution to support CKS with multiple query nodes. We also analyze our solution's effectiveness theoretically. Comprehensive experimental studies on various real-world datasets demonstrate our method's superiority.

翻译：根据一张G$G的图表,一个查询节点美元和一个整金美元,社区搜索(CS)从包含$q$的G$中寻找一个具有凝聚力的子集(用社区模型衡量,如美元-核心或美元-rues等社区模型衡量),它包含$q美元。对于对图形复杂性了解较少的普通用户来说,很难设定一个适当的K美元。即使我们定义了相当大的美元,CS返回的社区规模往往太大,用户无法深入了解。与整个社区相比,社区的关键成员似乎比其他社区更有价值。为此,我们专注于一个新的问题,即:textbf{C}omunity\ textbf{K}ey-members\ textbf{S}earch 问题。我们把观点转向社区中含有$quoality的的关键成员。为了解决CKS问题,我们首先提出基于tus decomplace的精确算法,作为基线。然后,我们展示了四个随机的Sload-rental-ralalalalalalalalalalal exaltial deal deal ex restialtial deal deal deal deal demo max maxism lautus lax lax lax lax lax lax lax lax lax laut ro ro max routus lautus laus lax max lautus max max lautd max lautus max max max max max max max max max max max max max max max max max max max max max max mas max max max max max max max max max max max 一种我们以我们以我们通过一个非常有效的最精度的精制的精度的精度方法,我们最精制的精度的精度算算算。我们最精度方法,我们最精度方法,我们最精度的