DRank+

A directory based PageRank prediction method for fast pagerank convergence

Hung-Yu Kao, Chia Sheng Liu, Yu Chuan Tsai, Chia Chun Shih, Tse Ming Tsai

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In recent years, most part of search engines use link analysis algorithms to measure the importance of web pages. The most famous link analysis algorithm is PageRank algorithm. However, previous researches in recent years have found that there exists an inherent bias against newly created pages in PageRank. In the previous work, a new ranking algorithm called DRank has been proposed to solve this issue. It utilizes the cluster phenomenon of PageRank in a directory to predict the possible importance of pages in the future and to diminish the inherent bias of search engines to new pages. In this paper, we modify the original DRank algorithm to complement the weaker part of DRank which could fail while the number of pages in directory is not enough. In our experiments, the augmented algorithm, i.e., DRank+ algorithm, obtains more accuracy in predicting the importance score of pages at next time stage than the original DRank algorithm. DRank+ not only alleviates the bias of newly created pages successfully but also reaches more accuracy than Page Quality and original DRank in predicting the importance of newly created pages.

Original languageEnglish
Title of host publicationWEBIST 2008 - 4th International Conference on Web Information Systems and Technologies, Proceedings
Pages175-180
Number of pages6
Volume2
Publication statusPublished - 2008
EventWEBIST 2008 - 4th International Conference on Web Information Systems and Technologies - Funchal, Madeira, Portugal
Duration: 2008 May 42008 May 7

Other

OtherWEBIST 2008 - 4th International Conference on Web Information Systems and Technologies
CountryPortugal
CityFunchal, Madeira
Period08-05-0408-05-07

Fingerprint

Search engines
Prediction
Drinks
PageRank
Websites
Experiments
Search engine
Link analysis
World Wide Web
Ranking
Experiment

All Science Journal Classification (ASJC) codes

  • Artificial Intelligence
  • Computer Networks and Communications
  • Computer Science Applications
  • Information Systems and Management

Cite this

Kao, H-Y., Liu, C. S., Tsai, Y. C., Shih, C. C., & Tsai, T. M. (2008). DRank+: A directory based PageRank prediction method for fast pagerank convergence. In WEBIST 2008 - 4th International Conference on Web Information Systems and Technologies, Proceedings (Vol. 2, pp. 175-180)
Kao, Hung-Yu ; Liu, Chia Sheng ; Tsai, Yu Chuan ; Shih, Chia Chun ; Tsai, Tse Ming. / DRank+ : A directory based PageRank prediction method for fast pagerank convergence. WEBIST 2008 - 4th International Conference on Web Information Systems and Technologies, Proceedings. Vol. 2 2008. pp. 175-180
@inproceedings{0ca9b0cc9d1542ceb6eba6375161ae2f,
title = "DRank+: A directory based PageRank prediction method for fast pagerank convergence",
abstract = "In recent years, most part of search engines use link analysis algorithms to measure the importance of web pages. The most famous link analysis algorithm is PageRank algorithm. However, previous researches in recent years have found that there exists an inherent bias against newly created pages in PageRank. In the previous work, a new ranking algorithm called DRank has been proposed to solve this issue. It utilizes the cluster phenomenon of PageRank in a directory to predict the possible importance of pages in the future and to diminish the inherent bias of search engines to new pages. In this paper, we modify the original DRank algorithm to complement the weaker part of DRank which could fail while the number of pages in directory is not enough. In our experiments, the augmented algorithm, i.e., DRank+ algorithm, obtains more accuracy in predicting the importance score of pages at next time stage than the original DRank algorithm. DRank+ not only alleviates the bias of newly created pages successfully but also reaches more accuracy than Page Quality and original DRank in predicting the importance of newly created pages.",
author = "Hung-Yu Kao and Liu, {Chia Sheng} and Tsai, {Yu Chuan} and Shih, {Chia Chun} and Tsai, {Tse Ming}",
year = "2008",
language = "English",
isbn = "9789898111265",
volume = "2",
pages = "175--180",
booktitle = "WEBIST 2008 - 4th International Conference on Web Information Systems and Technologies, Proceedings",

}

Kao, H-Y, Liu, CS, Tsai, YC, Shih, CC & Tsai, TM 2008, DRank+: A directory based PageRank prediction method for fast pagerank convergence. in WEBIST 2008 - 4th International Conference on Web Information Systems and Technologies, Proceedings. vol. 2, pp. 175-180, WEBIST 2008 - 4th International Conference on Web Information Systems and Technologies, Funchal, Madeira, Portugal, 08-05-04.

DRank+ : A directory based PageRank prediction method for fast pagerank convergence. / Kao, Hung-Yu; Liu, Chia Sheng; Tsai, Yu Chuan; Shih, Chia Chun; Tsai, Tse Ming.

WEBIST 2008 - 4th International Conference on Web Information Systems and Technologies, Proceedings. Vol. 2 2008. p. 175-180.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - DRank+

T2 - A directory based PageRank prediction method for fast pagerank convergence

AU - Kao, Hung-Yu

AU - Liu, Chia Sheng

AU - Tsai, Yu Chuan

AU - Shih, Chia Chun

AU - Tsai, Tse Ming

PY - 2008

Y1 - 2008

N2 - In recent years, most part of search engines use link analysis algorithms to measure the importance of web pages. The most famous link analysis algorithm is PageRank algorithm. However, previous researches in recent years have found that there exists an inherent bias against newly created pages in PageRank. In the previous work, a new ranking algorithm called DRank has been proposed to solve this issue. It utilizes the cluster phenomenon of PageRank in a directory to predict the possible importance of pages in the future and to diminish the inherent bias of search engines to new pages. In this paper, we modify the original DRank algorithm to complement the weaker part of DRank which could fail while the number of pages in directory is not enough. In our experiments, the augmented algorithm, i.e., DRank+ algorithm, obtains more accuracy in predicting the importance score of pages at next time stage than the original DRank algorithm. DRank+ not only alleviates the bias of newly created pages successfully but also reaches more accuracy than Page Quality and original DRank in predicting the importance of newly created pages.

AB - In recent years, most part of search engines use link analysis algorithms to measure the importance of web pages. The most famous link analysis algorithm is PageRank algorithm. However, previous researches in recent years have found that there exists an inherent bias against newly created pages in PageRank. In the previous work, a new ranking algorithm called DRank has been proposed to solve this issue. It utilizes the cluster phenomenon of PageRank in a directory to predict the possible importance of pages in the future and to diminish the inherent bias of search engines to new pages. In this paper, we modify the original DRank algorithm to complement the weaker part of DRank which could fail while the number of pages in directory is not enough. In our experiments, the augmented algorithm, i.e., DRank+ algorithm, obtains more accuracy in predicting the importance score of pages at next time stage than the original DRank algorithm. DRank+ not only alleviates the bias of newly created pages successfully but also reaches more accuracy than Page Quality and original DRank in predicting the importance of newly created pages.

UR - http://www.scopus.com/inward/record.url?scp=58049175923&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=58049175923&partnerID=8YFLogxK

M3 - Conference contribution

SN - 9789898111265

VL - 2

SP - 175

EP - 180

BT - WEBIST 2008 - 4th International Conference on Web Information Systems and Technologies, Proceedings

ER -

Kao H-Y, Liu CS, Tsai YC, Shih CC, Tsai TM. DRank+: A directory based PageRank prediction method for fast pagerank convergence. In WEBIST 2008 - 4th International Conference on Web Information Systems and Technologies, Proceedings. Vol. 2. 2008. p. 175-180