Automated detection of influential patents using singular values

Dohyun Kim, Bangrae Lee, Hyuck Jai Lee, Sang Pil Lee, Yeongho Moon, Myong-Kee Jeong

Research output: Contribution to journalArticle

10 Citations (Scopus)

Abstract

Centrality measures such as degree centrality have been utilized to identify influential and important patents in a citation network. However, no existing centrality measures take into consideration information from the change of the similarity matrix. This paper presents a new centrality measure based on the change of a node similarity matrix. The proposed approach gives more intuitive understanding of the finding of the influential nodes. The present study starts off with the assumption that the change of matrix that may result from removing a given node would assess the importance of the node since each node make a contribution to the given similarity matrix between nodes. The various matrix norms using the singular values such as nuclear norm which is the sum of all singular values, are used for calculating the contribution of a given node to a node similarity matrix. In other words, we can obtain the change of matrix norms for a given node after we calculate the singular values for the case of the nonexistence and the case of existence of the node. Then, the node resulting in the largest change (i.e., decrease) of matrix norms can be considered as the most important node. Computation of singular values can be computationally intensive when the similarity matrix size is large. Therefore, the singular value update technique is also developed for the case of the network with large nodes. We compare the performance of our proposed approach with other widely used centrality measures using U.S. patents data in the area of information and security. Experimental results show that our proposed approach is competitive or even performs better compared to existing approaches. Note to Practitioners - Identifying the influential patents manually by experts requires much time and efforts. Centrality measures in citation network have the advantage of providing the support for filtering such influential patents using an automated process. However, existing centrality techniques commonly suffer from the computational burden of the network with large nodes. To avoid this problem, this paper introduces a singular value based centrality which can be applied even to the case of the network with extremely large nodes. Especially, the new row updating and downdating techniques are proposed to obtain the singular values efficiently from a node similarity matrix of a large network. Our proposed approach makes it possible to find influential and important nodes efficiently in any large networks.

Original languageEnglish (US)
Article number6302211
Pages (from-to)723-733
Number of pages11
JournalIEEE Transactions on Automation Science and Engineering
Volume9
Issue number4
DOIs
StatePublished - Sep 21 2012

All Science Journal Classification (ASJC) codes

  • Control and Systems Engineering
  • Electrical and Electronic Engineering

Keywords

  • Acyclic directed networks
  • centrality
  • citation network
  • singular value decomposition
  • social network

Cite this

Kim, Dohyun ; Lee, Bangrae ; Lee, Hyuck Jai ; Lee, Sang Pil ; Moon, Yeongho ; Jeong, Myong-Kee. / Automated detection of influential patents using singular values. In: IEEE Transactions on Automation Science and Engineering. 2012 ; Vol. 9, No. 4. pp. 723-733.
@article{3a95bb9ad65c426593d7af9d68000bc2,
title = "Automated detection of influential patents using singular values",
abstract = "Centrality measures such as degree centrality have been utilized to identify influential and important patents in a citation network. However, no existing centrality measures take into consideration information from the change of the similarity matrix. This paper presents a new centrality measure based on the change of a node similarity matrix. The proposed approach gives more intuitive understanding of the finding of the influential nodes. The present study starts off with the assumption that the change of matrix that may result from removing a given node would assess the importance of the node since each node make a contribution to the given similarity matrix between nodes. The various matrix norms using the singular values such as nuclear norm which is the sum of all singular values, are used for calculating the contribution of a given node to a node similarity matrix. In other words, we can obtain the change of matrix norms for a given node after we calculate the singular values for the case of the nonexistence and the case of existence of the node. Then, the node resulting in the largest change (i.e., decrease) of matrix norms can be considered as the most important node. Computation of singular values can be computationally intensive when the similarity matrix size is large. Therefore, the singular value update technique is also developed for the case of the network with large nodes. We compare the performance of our proposed approach with other widely used centrality measures using U.S. patents data in the area of information and security. Experimental results show that our proposed approach is competitive or even performs better compared to existing approaches. Note to Practitioners - Identifying the influential patents manually by experts requires much time and efforts. Centrality measures in citation network have the advantage of providing the support for filtering such influential patents using an automated process. However, existing centrality techniques commonly suffer from the computational burden of the network with large nodes. To avoid this problem, this paper introduces a singular value based centrality which can be applied even to the case of the network with extremely large nodes. Especially, the new row updating and downdating techniques are proposed to obtain the singular values efficiently from a node similarity matrix of a large network. Our proposed approach makes it possible to find influential and important nodes efficiently in any large networks.",
keywords = "Acyclic directed networks, centrality, citation network, singular value decomposition, social network",
author = "Dohyun Kim and Bangrae Lee and Lee, {Hyuck Jai} and Lee, {Sang Pil} and Yeongho Moon and Myong-Kee Jeong",
year = "2012",
month = "9",
day = "21",
doi = "10.1109/TASE.2012.2210214",
language = "English (US)",
volume = "9",
pages = "723--733",
journal = "IEEE Transactions on Automation Science and Engineering",
issn = "1545-5955",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "4",

}

Automated detection of influential patents using singular values. / Kim, Dohyun; Lee, Bangrae; Lee, Hyuck Jai; Lee, Sang Pil; Moon, Yeongho; Jeong, Myong-Kee.

In: IEEE Transactions on Automation Science and Engineering, Vol. 9, No. 4, 6302211, 21.09.2012, p. 723-733.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Automated detection of influential patents using singular values

AU - Kim, Dohyun

AU - Lee, Bangrae

AU - Lee, Hyuck Jai

AU - Lee, Sang Pil

AU - Moon, Yeongho

AU - Jeong, Myong-Kee

PY - 2012/9/21

Y1 - 2012/9/21

N2 - Centrality measures such as degree centrality have been utilized to identify influential and important patents in a citation network. However, no existing centrality measures take into consideration information from the change of the similarity matrix. This paper presents a new centrality measure based on the change of a node similarity matrix. The proposed approach gives more intuitive understanding of the finding of the influential nodes. The present study starts off with the assumption that the change of matrix that may result from removing a given node would assess the importance of the node since each node make a contribution to the given similarity matrix between nodes. The various matrix norms using the singular values such as nuclear norm which is the sum of all singular values, are used for calculating the contribution of a given node to a node similarity matrix. In other words, we can obtain the change of matrix norms for a given node after we calculate the singular values for the case of the nonexistence and the case of existence of the node. Then, the node resulting in the largest change (i.e., decrease) of matrix norms can be considered as the most important node. Computation of singular values can be computationally intensive when the similarity matrix size is large. Therefore, the singular value update technique is also developed for the case of the network with large nodes. We compare the performance of our proposed approach with other widely used centrality measures using U.S. patents data in the area of information and security. Experimental results show that our proposed approach is competitive or even performs better compared to existing approaches. Note to Practitioners - Identifying the influential patents manually by experts requires much time and efforts. Centrality measures in citation network have the advantage of providing the support for filtering such influential patents using an automated process. However, existing centrality techniques commonly suffer from the computational burden of the network with large nodes. To avoid this problem, this paper introduces a singular value based centrality which can be applied even to the case of the network with extremely large nodes. Especially, the new row updating and downdating techniques are proposed to obtain the singular values efficiently from a node similarity matrix of a large network. Our proposed approach makes it possible to find influential and important nodes efficiently in any large networks.

AB - Centrality measures such as degree centrality have been utilized to identify influential and important patents in a citation network. However, no existing centrality measures take into consideration information from the change of the similarity matrix. This paper presents a new centrality measure based on the change of a node similarity matrix. The proposed approach gives more intuitive understanding of the finding of the influential nodes. The present study starts off with the assumption that the change of matrix that may result from removing a given node would assess the importance of the node since each node make a contribution to the given similarity matrix between nodes. The various matrix norms using the singular values such as nuclear norm which is the sum of all singular values, are used for calculating the contribution of a given node to a node similarity matrix. In other words, we can obtain the change of matrix norms for a given node after we calculate the singular values for the case of the nonexistence and the case of existence of the node. Then, the node resulting in the largest change (i.e., decrease) of matrix norms can be considered as the most important node. Computation of singular values can be computationally intensive when the similarity matrix size is large. Therefore, the singular value update technique is also developed for the case of the network with large nodes. We compare the performance of our proposed approach with other widely used centrality measures using U.S. patents data in the area of information and security. Experimental results show that our proposed approach is competitive or even performs better compared to existing approaches. Note to Practitioners - Identifying the influential patents manually by experts requires much time and efforts. Centrality measures in citation network have the advantage of providing the support for filtering such influential patents using an automated process. However, existing centrality techniques commonly suffer from the computational burden of the network with large nodes. To avoid this problem, this paper introduces a singular value based centrality which can be applied even to the case of the network with extremely large nodes. Especially, the new row updating and downdating techniques are proposed to obtain the singular values efficiently from a node similarity matrix of a large network. Our proposed approach makes it possible to find influential and important nodes efficiently in any large networks.

KW - Acyclic directed networks

KW - centrality

KW - citation network

KW - singular value decomposition

KW - social network

UR - http://www.scopus.com/inward/record.url?scp=84867404121&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84867404121&partnerID=8YFLogxK

U2 - 10.1109/TASE.2012.2210214

DO - 10.1109/TASE.2012.2210214

M3 - Article

VL - 9

SP - 723

EP - 733

JO - IEEE Transactions on Automation Science and Engineering

JF - IEEE Transactions on Automation Science and Engineering

SN - 1545-5955

IS - 4

M1 - 6302211

ER -