Calculating level-based importance of a web page

• US 20070198504A1
• Filed: 02/23/2006
• Published: 08/23/2007
• Est. Priority Date: 02/23/2006
• Status: Abandoned Application

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Ma, Wei-Ying, Liu, Tie-Yan, Feng, Guang
• Application Number:
  US11/360,987
• Publication Number:
  US 20070198504A1
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques

Abstract

A method and system for determining importance of web pages that factors in the level of the web page within a web site hierarchy is provided. The importance system calculates the importance of web pages based on links between web pages. The importance system calculates a weight for a link between a from web page and a to web page based on the level of the from web page within its web site hierarchy. The importance system may use various algorithms for calculating the importance of web pages that factor in the weights of the links. The importance system may also factor in the level of a to web page within a web site hierarchy when calculating the weight of a link between a from web page and the to web page.

Claims

1. A system for determining importance of pages, comprising:

a component that determines a weight of a link between a from page and a to page, the weight being based on a level of the from page within a hierarchy of pages that contains the from page; and

a component that calculates importance of a page based on the weights of links from from pages to the page.

2. The system of claim 1 wherein the weight is calculated according to the following equation:

${\tilde{w}}_{\mathrm{wj}|i}=\frac{1}{l_i·<br><br>l_{\mathrm{anc}<br><br>\left(i,j\right)}}$

where {tilde over (w)}_wj|i is the weight of a link from page i to page j, l_i is the level of page i, and l_anc(i,j) is the level of the closest common ancestor page of page i and page j.

3. The system of claim 1 wherein the weight varies non-linearly based on level.

4. The system of claim 1 wherein the weight is a linear weight that is biased based on level.

5. The system of claim 1 wherein the importance is calculated based on an algorithm that factors in hub and authority scores of pages.

6. The system of claim 5 where the importance is based on the following equation:

$a^{\left(t+1\right)}={\tilde{L}}^T<br><br>h^{\left(t\right)}=\left({\tilde{L}}^T<br><br>\tilde{L}\right)<br><br>a^{\left(t\right)}$ $h^{\left(t+1\right)}=\tilde{L}<br><br>\text{\hspace{1em}<br><br>}<br><br>a^{\left(t\right)}=\left(\tilde{L}<br><br>\text{\hspace{1em}<br><br>}<br><br>{\tilde{L}}^T\right)<br><br>h^{\left(t\right)}$

where a is a vector of authority scores of pages, h is a vector of hub scores of pages, and {tilde over (L)} is a matrix of weights of links between pairs of pages.

7. The system of claim 1 wherein the importance is calculated based on a page rank algorithm with the weights as values of an adjacency matrix.

8. The system of claim 1 wherein the calculated importance of a page factors in the level of the page.

9. The system of claim 8 wherein the calculated importance of a page decreases as its depth within a hierarchy increases.

10. The system of claim 1

wherein the weight is calculated according to the following equation:

${\tilde{w}}_{j|i}=\frac{1}{l_i·<br><br>l_{\mathrm{anc}<br><br>\left(i,j\right)}}$

where w_j|i is the weight of a link from page i to page j, l_i is the level of page i, and l_anc(i,j) is the level of the closest common ancestor page of page i and page j;

wherein the importance is based on the following equation;

$a^{\left(t+1\right)}={\tilde{L}}^T<br><br>h^{\left(t\right)}=\left({\tilde{L}}^T<br><br>\tilde{L}\right)<br><br>a^{\left(t\right)}$$h^{\left(t+1\right)}=\tilde{L}<br><br>a^T=\left(\tilde{L}<br><br>{\tilde{L}}^T\right)<br><br>h^{\left(t\right)}$

where a is a vector of authority scores of pages, h is a vector of hub scores of pages, and {tilde over (L)} is a matrix of weights of links between pairs of pages; and

wherein the calculated importance of a page decreases as its depth within a hierarchy increases.

11. A computer-readable medium containing instructions for controlling a computing device to determine weights of links between web pages, by a method comprising:

providing levels of web pages within web sites; and

calculating weights of links between from web pages and to web pages based on the levels of the from web pages and the levels of the to web pages.

12. The computer-readable medium of claim 11 wherein the weights are calculated according to the following equation:

$w_{j|i}=\frac{1}{l_i·<br><br>l_j·<br><br>l_{\mathrm{anc}<br><br>\left(i,j\right)}}$

where w_j|iis the weight of a link from page i to page j, l_i is the level of page i, l_j is the level of web page l, and l_anc(i,j) is the level of the closest common ancestor page of page i and page j.

13. The computer-readable medium of claim 11 wherein the weights of links decrease as the depths of the to web pages within a hierarchy increase.

14. The computer-readable medium of claim 11 wherein the weights of links increase as the depths of the from web pages within a hierarchy decrease.

15. The computer-readable medium of claim 11 including calculating importance of web pages based on the calculated weights of links between from web pages and to web pages.

16. The computer-readable medium of claim 15 wherein the importance is calculated based on an algorithm that factors in hub and authority scores of web pages.

17. A system for determining importance of web pages, comprising:

a component that calculates importance of web pages based on links between from web pages and to web pages wherein the calculated importance of a web page decreases as its depth within a hierarchy increases.

18. The system of claim 17 wherein the importance is calculated based on an algorithm that factors in hub and authority scores of web pages.

19. The system of claim 17 wherein the importance of a web page increases as the depth of a web page within a hierarchy with a link to it decreases.

20. The system of claim 18 wherein the importance is calculated based on an algorithm that factors in hub and authority scores of web pages.