scispace - formally typeset
Search or ask a question
Topic

SimRank

About: SimRank is a research topic. Over the lifetime, 250 publications have been published within this topic receiving 21163 citations.


Papers
More filters
Proceedings ArticleDOI
14 Jun 2015
TL;DR: This paper firstly builds the co-purchasing network by using the relationships between different type products, and then compute the similarity between products using SimRank, and gives some experimental results by implementing this method on Amazon dataset.
Abstract: Online bookstores have attracted millions of people and helped provide them hopeful books. Similarity search over on-line book store mainly focuses on finding the top-K most similar products for a given query. In this paper, we discuss how to find similar products for a given query product, and propose a framework for finding similar products from online bookstore. We firstly build the co-purchasing network by using the relationships between different type products, and then compute the similarity between products using SimRank. Finally, we give some experimental results by implementing this method on Amazon dataset, which demonstrate that the proposed method can find the underlying results over real dataset.
Book ChapterDOI
01 Jan 2020
TL;DR: This chapter shows the application of mining social network graphs on the tidal and wave energy system by a further application of clique percolation and SimRank implementation.
Abstract: The increasing popularity of social networks is clearly demonstrated by the huge number of users acquired in a short amount of time. This chapter shows the application of mining social network graphs on the tidal and wave energy system. At the primary level ocean energy systems are assessed through social networks by a further application of clique percolation and SimRank implementation.
Posted Content
TL;DR: Li et al. as mentioned in this paper proposed a memory-efficient algorithm for all-pairs SimRank, which requires only O(kn+m) memory and O(n^2) memory in the worst case.
Abstract: In this article, we study the efficient dynamical computation of all-pairs SimRanks on time-varying graphs. Li {\em et al}.'s approach requires $O(r^4 n^2)$ time and $O(r^2 n^2)$ memory in a graph with $n$ nodes, where $r$ is the target rank of the low-rank SVD. (1) We first consider edge update that does not accompany new node insertions. We show that the SimRank update $\Delta S$ in response to every link update is expressible as a rank-one Sylvester matrix equation. This provides an incremental method requiring $O(Kn^2)$ time and $O(n^2)$ memory in the worst case to update all pairs of similarities for $K$ iterations. (2) To speed up the computation further, we propose a lossless pruning strategy that captures the "affected areas" of $\Delta S$ to eliminate unnecessary retrieval. This reduces the time of the incremental SimRank to $O(K(m+|AFF|))$, where $m$ is the number of edges in the old graph, and $|AFF| (\le n^2)$ is the size of "affected areas" in $\Delta S$, and in practice, $|AFF| \ll n^2$. (3) We also consider edge updates that accompany node insertions, and categorize them into three cases, according to which end of the inserted edge is a new node. For each case, we devise an efficient incremental algorithm that can support new node insertions. (4) We next design an efficient batch incremental method that handles "similar sink edges" simultaneously and eliminates redundant edge updates. (5) To achieve linear memory, we devise a memory-efficient strategy that dynamically updates all pairs of SimRanks column by column in just $O(Kn+m)$ memory, without the need to store all $(n^2)$ pairs of old SimRank scores. Experimental studies on various datasets demonstrate that our solution substantially outperforms the existing incremental SimRank methods, and is faster and more memory-efficient than its competitors on million-scale graphs.
Journal ArticleDOI
TL;DR: By using the link prediction algorithm based on the similarity, this paper tried to predict the links’ connection relationship and got the possible network topology structure among the network’ ASes when the actual connection link don’t be completely detected.
Abstract: Multiple Autonomous systems (ASes) of the network are usually consisted of many routers such as inter-router and intra-router connection each other. In order to detect the network routers connection relationship, many researchers at home or abroad use various methods such as active end-to-end links detection or observing SNMP MIBs to understand the links’ connection relationship among the routers in network’ ASes. In this paper, by using the link prediction algorithm based on the similarity, we tried to predict the links’ connection relationship and got the possible network topology structure among the network’ ASes when the actual connection link don’t be completely detected. Through the experiments, the prediction results can be seen that the similarity algorithm index of ACT, SRW, and SimRank based on the random walk can also achieve better prediction accuracy above 0.95, which prove the similarity index especially the random walk algorithm can realize the higher link prediction accuracy to the network’ ASes under the conditions of missing some known connection link, etc.
Patent
27 Sep 2019
TL;DR: In this paper, the authors proposed a collaborative filtering recommendation method based on a single source SimRank, which comprises the steps of converting to-be-recommended users, users and relationships among the users into graph structures, according to the probability Pil (u, u) that a random walk starting from a source node u arrives at a node w through a step 1 and stops, wherein the probability eta (w) is that two random walks starting from the node w do not meet each other in the walking process, and the probability PT (v, w) was that a
Abstract: The invention discloses a collaborative filtering recommendation method based on a single source SimRank. The method comprises the steps of converting to-be-recommended users, users and relationships among the users into graph structures, according to the probability Pil (u, u) that a random walk starting from a source node u arrives at a node w through a step 1 and stops, wherein the probability eta (w) is that two random walks starting from the node w do not meet each other in the walking process, and the probability Pil (v, w) is that a reverse walk starting from the node w reaches the node v and stops after passing through the step l; estimating the SimRank similarity between the nodes u and v according to the SimRank similarity between the nodes u and v, and repeatedly executing the similarity estimation until the estimation between all the nodes in the graph structure and the source node u is completed; finding the first k nodes with the highest similarity with the to-be-recommended node according to the estimation result; and obtaining behavior information of the first k nodes, and integrating and pushing the behavior information to the source node u. According to the collaborative filtering recommendation method based on the single-source SimRank provided by the embodiment of the invention, the time complexity is reduced, and the requirements of real-time recommendation and interactive query are met.

Network Information
Related Topics (5)
Web page
50.3K papers, 975.1K citations
77% related
Graph (abstract data type)
69.9K papers, 1.2M citations
77% related
Ontology (information science)
57K papers, 869.1K citations
77% related
Scalability
50.9K papers, 931.6K citations
74% related
Tree (data structure)
44.9K papers, 749.6K citations
74% related
Performance
Metrics
No. of papers in the topic in previous years
YearPapers
202115
202026
201916
201817
201719
201616