Local approximation of pagerank and reverse pagerank

Bar-Yossef, Ziv; Mashiach, Li-Tal

doi:10.1145/1458082.1458122

Cited by 48 publications

(29 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As we have seen in the previous subsections, PageRank can be adapted to weighted digraphs and supports personalization. Additionally, it can be efficiently approximated [2,5], and can be computed in a parallel or distributed framework [31,44].…”

Section: Pagerank In Contextmentioning

confidence: 99%

Endorsement deduction and ranking in social networks

Pérez-Rosés

Sebé

Ribó

2016

Computer Communications

View full text Add to dashboard Cite

Some social networks, such as LinkedIn and ResearchGate, allow user endorsements for specific skills. In this way, for each skill we get a directed graph where the nodes correspond to users' profiles and the arcs represent endorsement relations. From the number and quality of the endorsements received, an authority score can be assigned to each profile. In this paper we propose an authority score computation method that takes into account the relations existing among different skills. Our method is based on enriching the information contained in the digraph of endorsements corresponding to a specific skill, and then applying a ranking method admitting weighted digraphs, such as PageRank. We describe the method, and test it on a synthetic network of 1493 nodes, fitted with endorsements.

show abstract

Section: Pagerank In Contextmentioning

confidence: 99%

Endorsement deduction and ranking in social networks

Pérez-Rosés

Sebé

Ribó

2016

Computer Communications

View full text Add to dashboard Cite

show abstract

“…where [13]. Given two nodes u and v, v's social influence on u is realized along the different paths from v to u in the graph.…”

Section: ) Computation Cost Analysismentioning

confidence: 99%

Semantic-Based Recommendation Across Heterogeneous Domains

Yang

Song

Wang

2015

2015 IEEE International Conference on Data Mining

View full text Add to dashboard Cite

Abstract-Cross-domain recommendation has attracted wide research interest which generally aims at improving the recommendation performance by alleviating the cold start problem in collaborative filtering based recommendation or generating a more comprehensive user profiles from multiple domains. In most previous cross-domain recommendation settings, explicit or implicit relationships can be easily established across different domains. However, many real applications belong to a more challenging setting: recommendation across heterogeneous domains without explicit relationships, where neither explicit user-item relations nor overlapping features exist between different domains. In this new setting, we need to (1) enrich the sparse data to characterize users or items and (2) bridge the gap caused by the heterogenous features in different domains. To overcome the first challenge, we proposed an optimized local tag propagation algorithm to generate descriptive tags for user profiling. For the second challenge, we proposed a semantic relatedness metric by mapping the heterogenous features onto their concept space derived from online encyclopedias. We conducted extensive experiments on two real datasets to justify the effectiveness of our solution.

show abstract

“…The PPR from node i to j is computed as P P R(i, j) = mj/N . Our experiments use α = 0.15, the common choice of PPR computations [18,5], and T = log n/α because prior work proved that PageRank converges in O(log n/α) hops [11]. We also found that N = 8000 is adequate to get a stable PPR estimation.…”

Section: Ground Truth Computationmentioning

confidence: 99%

On the embeddability of random walk distances

et al. 2013

View full text Add to dashboard Cite

Analysis of large graphs is critical to the ongoing growth of search engines and social networks. One class of queries centers around node affinity, often quantified by random-walk distances between node pairs, including hitting time, commute time, and personalized PageRank (PPR). Despite the potential of these "metrics," they are rarely, if ever, used in practice, largely due to extremely high computational costs.In this paper, we investigate methods to scalably and efficiently compute random-walk distances, by "embedding" graphs and distances into points and distances in geometric coordinate spaces. We show that while existing graph coordinate systems (GCS) can accurately estimate shortest path distances, they produce significant errors when embedding random-walk distances. Based on our observations, we propose a new graph embedding system that explicitly accounts for per-node graph properties that affect random walk. Extensive experiments on a range of graphs show that our new approach can accurately estimate both symmetric and asymmetric random-walk distances. Once a graph is embedded, our system can answer queries between any two nodes in 8 microseconds, orders of magnitude faster than existing methods. Finally, we show that our system produces estimates that can replace ground truth in applications with minimal impact on application output.

show abstract

Local approximation of pagerank and reverse pagerank

Cited by 48 publications

References 35 publications

Endorsement deduction and ranking in social networks

Endorsement deduction and ranking in social networks

Semantic-Based Recommendation Across Heterogeneous Domains

On the embeddability of random walk distances

Contact Info

Product

Resources

About