PageRank revisited

We consider the problem of approximating the PageRank of a target node using only local information provided by a link server. This problem was originally studied by Chen, Gan, and Suel (CIKM 2004), who presented an algorithm for tackling it. We prove that local approximation of PageRank, even to within modest approximation factors, is infeasible in the worst-case, as it requires probing the link server for Ω(n) nodes, where n is the size of the graph. The difficulty emanates from nodes of high in-degree and/or from slow convergence of the PageRank random walk.We show that when the graph has bounded in-degree and admits fast PageRank convergence, then local PageRank approximation can be done using a small number of queries. Unfortunately, natural graphs, such as the web graph, are abundant with high in-degree nodes, making this algorithm (or any other local approximation algorithm) too costly. On the other hand, reverse natural graphs tend to have low in-degree while maintaining fast PageRank convergence. It follows that calculating Reverse PageRank locally is frequently more feasible than computing PageRank locally.We demonstrate that Reverse PageRank is useful for several applications, including computation of hub scores for web pages, finding influencers in social networks, obtaining good seeds for crawling, and measurement of semantic relatedness between concepts in a taxonomy.

show abstract

“…For more information about PageRank, refer to recent surveys on the topic [4,5]. Local PageRank approximation.…”

Section: Preliminariesmentioning

confidence: 99%

Local approximation of PageRank and reverse PageRank

Bar-Yossef

Mashiach

2008

Proceedings of the 31st Annual International ACM SIGIR Conference on Research and Development in Information Retrieval

View full text Add to dashboard Cite

show abstract

“…, will give us the distribution vector of random web user after steps. This kind of nature is known as an example of Markov method or Markov matrix [14,25]. It is clear that for any starting PageRank vector, the markov method converges to a unique stationary vector when matrix M is stochastic and primitive i.e.…”

Section: Basic Pagerank Modelmentioning

confidence: 99%

“…(7) [8,25]: (7) Where is the teleportation matrix of , controls priority given to the hyperlink structure over artificial teleportation matrix . Many researchers state that the value of damping factor affect the convergence rate of PageRank Power method [11,12,14,18,19]. We have performed an experiment to observe the impact of damping factor on the convergence rate of PageRank algorithm, and ordering of web pages displayed on web search engine on various datasets by using following Eq.…”

Section: Effect Of Damping Factor On Pagerank Algorithmmentioning

confidence: 99%

“…In initial approximation, we could define PageRank as the ratio of [8,11,12,13]. To resolve rank sink issue, at every step the random web user has the choice to choose any out-link with probability , and will restart from another node of n web pages chosen at random with probability , where is a damping factor [7,14]. Brin and Page, the founders of Google web search engine, suggest to take the value of damping factor =0.85.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Discussion on Damping Factor Value in PageRank Computation

Srivastava¹,

Garg²,

Mishra³

2017

IJISA

View full text Add to dashboard Cite

Abstract-Web search engines use various ranking methods to determine the order of web pages displayed on the Search Engine Result Page (SERP). PageRank is one of the popular and widely used ranking method. PageRank of any web page can be defined as a fraction of time a random web surfer spends on that web page on average. The PageRank method is a stationary distribution of a stochastic method whose states are web pages of the Web graph. This stochastic method is acquired by combining the hyperlink matrix of the web graph and a trivial uniform process. This combination is needed to make primitive so that stationary distribution is well defined. The combination depends on the value of damping factor in the computation of PageRank. The damping factor parameter state that how much time random web surfer follow hyperlink structure than teleporting. The value of is exceptionally empirical and in current scenario = 0.85 is considered as suggested by Brin and Page. If we take α =0.8 then we can say that out of total time, 80% of time is taken by the random web surfer to follow the hyperlink structure and 20% time they teleport to new web pages randomly. Today web surfer gets worn out too early on the web because of non-availability of relevant information and they can easily teleport to new web pages rather than following hyperlink structure. So we have to choose some value of damping factor other than 0.85. In this paper, we have given an experimental analysis of PageRank computation for different value of the damping factor. We have observed that for value of =0.7, PageRank method takes fewer numbers of iterations to converge than =0.85, and for these values of the top 25 web pages returned by PageRank method in the SERP are almost same, only some of them exchange their positions. From the experimental results it is observed that value of damping factor =0.7 takes approximate 25-30% fewer numbers of iterations than =0.85 to get closely identical web pages in top 25 result pages for personalized web search, selective crawling, intra-web search engine.

show abstract

“…3) in the long term. When embedding one VNR each time, the VNE-TAGRD adopts a novel node-ranking approach, similar to the PageRank approach in web-searching area [23][24][25], to rank all substrate nodes and virtual nodes ahead, according to three essential topology attributes and global network resources. The greedy node mapping approach, based on the novel noderanking values, is then performed, with fulfilling the node constraints (virtual node location and virtual node capacity demands are considered in our paper).…”

Section: Introductionmentioning

confidence: 99%

A Novel Optimal Mapping Algorithm With Less Computational Complexity for Virtual Network Embedding

Cao

Zhu

Zheng

et al. 2018

IEEE Trans. Netw. Serv. Manage.

View full text Add to dashboard Cite

A novel optimal mapping algorithm with less computational complexity for virtual network embedding Abstract-Virtual Network Embedding (VNE) problem has been widely accepted as an important aspect in Network Virtualization (NV) area: how to efficiently embed virtual networks, with node and link resource demands, onto the shared substrate network that has finite network resources. Previous VNE heuristic algorithms, only considering single network topology attribute and local resources of each node, may lead to inefficient resource utilization of the substrate network in the long term. To address this issue, a Topology Attribute and Global Resource-Driven VNE algorithm (VNE-TAGRD), adopting a novel node-ranking approach, is proposed in this paper. The novel node-ranking approach, developed from the well-known Google PageRank algorithm, considers three essential topology attributes and global network resources information before conducting the embedding of given virtual network request (VNR). Numerical simulation results reveal that the VNE-TAGRD algorithm outperforms five typical and latest heuristic algorithms that only consider single network topology attribute and local resources of each node, such as long-term average VNR acceptance ratio and average revenue to cost ratio.

show abstract

PageRank revisited

Cited by 57 publications

References 3 publications

Local approximation of PageRank and reverse PageRank

Local approximation of PageRank and reverse PageRank

Discussion on Damping Factor Value in PageRank Computation

A Novel Optimal Mapping Algorithm With Less Computational Complexity for Virtual Network Embedding

Contact Info

Product

Resources

About