Main.HomePage History

Hide minor edits - Show changes to markup

May 04, 2010, at 11:27 AM by 213.27.241.137 -
Changed lines 3-4 from:

The goal of the Web Spam Challenge series to identify and compare Machine Learning (ML) methods for automatically labeling structured data represented as graphs. More precisely, we focus on the problem of labeling all nodes of a graph from a partial labeling of them. The application we study is Web Spam Detection, where we want to detect deliberate actions of deception aimed at the ranking functions used by search engines.

to:

The goal of the Web Spam Challenge series is to identify and compare Machine Learning (ML) methods for automatically labeling structured data represented as graphs. More precisely, we focus on the problem of labeling all nodes of a graph from a partial labeling of them. The application we study is Web Spam Detection, where we want to detect deliberate actions of deception aimed at the ranking functions used by search engines.

Added lines 9-12:

See also:

Discovery Challenge 2010

May 08, 2008, at 12:52 PM by 209.131.62.115 -
Deleted lines 4-5:
Start here: get the WEBSPAM-UK2007 dataset and optionally the pre-computed feature vectors.
March 21, 2008, at 09:36 AM by 89.6.43.167 -
Changed lines 3-4 from:

The goal of the Web Spam Challenge series to identify and compare Machine Learning (ML) methods for automatically labelling structured data represented as graphs. More precisely, we focus on the problem of labelling all nodes of a graph from a partial labelling of them. The application we study is Web Spam Detection, where we want to detect deliberate actions of deception aimed at the ranking functions used by search engines.

to:

The goal of the Web Spam Challenge series to identify and compare Machine Learning (ML) methods for automatically labeling structured data represented as graphs. More precisely, we focus on the problem of labeling all nodes of a graph from a partial labeling of them. The application we study is Web Spam Detection, where we want to detect deliberate actions of deception aimed at the ranking functions used by search engines.

March 17, 2008, at 07:16 AM by 84.88.76.49 -
Changed lines 5-6 from:
Start here: first download the WEBSPAM-UK2007 dataset and optionally the pre-computed feature vectors.
to:
Start here: get the WEBSPAM-UK2007 dataset and optionally the pre-computed feature vectors.
March 17, 2008, at 07:16 AM by 84.88.76.49 -
Changed lines 5-6 from:
How to participate in the Web Spam Challenge 2008: first download the WEBSPAM-UK2007 dataset and optionally the pre-computed feature vectors. <-
to:
Start here: first download the WEBSPAM-UK2007 dataset and optionally the pre-computed feature vectors.
March 17, 2008, at 07:15 AM by 84.88.76.49 -
Changed lines 5-6 from:
How to participate in the Web Spam Challenge 2008: first download the WEBSPAM-UK2007 dataset and optionally the pre-computed feature vectors.
to:
How to participate in the Web Spam Challenge 2008: first download the WEBSPAM-UK2007 dataset and optionally the pre-computed feature vectors. <-
March 17, 2008, at 07:15 AM by 84.88.76.49 -
Changed lines 5-6 from:
  • How to participate in the Web Spam Challenge 2008: first download the WEBSPAM-UK2007 dataset and optionally the pre-computed feature vectors.
to:
How to participate in the Web Spam Challenge 2008: first download the WEBSPAM-UK2007 dataset and optionally the pre-computed feature vectors.
March 17, 2008, at 07:15 AM by 84.88.76.49 -
March 17, 2008, at 07:15 AM by 84.88.76.49 -
Changed lines 5-6 from:

How to participate in the Web Spam Challenge 2008: first download the WEBSPAM-UK2007 dataset and optionally the pre-computed feature vectors.

to:
  • How to participate in the Web Spam Challenge 2008: first download the WEBSPAM-UK2007 dataset and optionally the pre-computed feature vectors.
March 12, 2008, at 05:46 AM by 84.88.76.49 -
Changed lines 5-6 from:

How to participate in the Web Spam Challenge 2008: first download the WEBSPAM-UK2007 dataset and the pre-computed feature vectors.

to:

How to participate in the Web Spam Challenge 2008: first download the WEBSPAM-UK2007 dataset and optionally the pre-computed feature vectors.

March 12, 2008, at 05:46 AM by 84.88.76.49 -
Changed lines 5-6 from:

How to participate in the Web Spam Challenge 2008: download the WEBSPAM-UK2007 dataset and some pre-computed feature vectors.

to:

How to participate in the Web Spam Challenge 2008: first download the WEBSPAM-UK2007 dataset and the pre-computed feature vectors.

March 12, 2008, at 05:46 AM by 84.88.76.49 -
Changed lines 5-6 from:

How to participate in the Web Spam Challenge 2008: download the WEBSPAM-UK2007 dataset and optionally some pre-computed feature vectors.

to:

How to participate in the Web Spam Challenge 2008: download the WEBSPAM-UK2007 dataset and some pre-computed feature vectors.

March 12, 2008, at 05:45 AM by 84.88.76.49 -
Changed lines 5-6 from:

How to participate in the Web Spam Challenge 2008: download the WEBSPAM-UK2007 dataset and optionally some pre-computed feature vectors (available in Matlab and ARFF format).

to:

How to participate in the Web Spam Challenge 2008: download the WEBSPAM-UK2007 dataset and optionally some pre-computed feature vectors.

March 12, 2008, at 05:39 AM by 84.88.76.49 -
Changed lines 5-6 from:

How to participate in the Web Spam Challenge 2008: download the WEBSPAM-UK2007 dataset and optionally some pre-computed feature vectors (available in Matlab and ARFF format).

to:

How to participate in the Web Spam Challenge 2008: download the WEBSPAM-UK2007 dataset and optionally some pre-computed feature vectors (available in Matlab and ARFF format).

March 12, 2008, at 05:39 AM by 84.88.76.49 -
Changed lines 5-6 from:

The Web Spam Challenge 2008 is open: download the WEBSPAM-UK2007 dataset and optionally some pre-computed feature vectors (available in Matlab and ARFF format).

to:

How to participate in the Web Spam Challenge 2008: download the WEBSPAM-UK2007 dataset and optionally some pre-computed feature vectors (available in Matlab and ARFF format).

March 12, 2008, at 05:39 AM by 84.88.76.49 -
Changed lines 5-6 from:

Getting started: download the WEBSPAM-UK2007 dataset and optionally some pre-computed feature vectors (available in Matlab and ARFF format).

to:

The Web Spam Challenge 2008 is open: download the WEBSPAM-UK2007 dataset and optionally some pre-computed feature vectors (available in Matlab and ARFF format).

March 12, 2008, at 05:38 AM by 84.88.76.49 -
Changed lines 5-6 from:

Getting started: download the WEBSPAM-UK2007 dataset and optionally some pre-computed feature vectors (available in Matlab and ARFF format).

to:

Getting started: download the WEBSPAM-UK2007 dataset and optionally some pre-computed feature vectors (available in Matlab and ARFF format).

March 12, 2008, at 05:38 AM by 84.88.76.49 -
Deleted lines 6-11:

2008 Timeline

  • 18 January 2008 - WEBSPAM-UK2007 dataset available
  • 6 March 2008 - Pre-computed feature sets are available
  • 13 April 2008 - Submission deadline for the challenge
March 12, 2008, at 05:36 AM by 84.88.76.49 -
March 12, 2008, at 05:36 AM by 84.88.76.49 -
Changed lines 5-6 from:

Where to start? Just download the WEBSPAM-UK2007 dataset and optionally some pre-computed feature vectors (available in Matlab and ARFF format).

to:

Getting started: download the WEBSPAM-UK2007 dataset and optionally some pre-computed feature vectors (available in Matlab and ARFF format).

Changed lines 13-16 from:

Mailing list

If you are interested in participating in a future Web Spam Challenge, please subscribe to our mailing list.

to:

Mailing list

If you are interested in the Web Spam Challenge, please subscribe to our mailing list.

March 12, 2008, at 05:34 AM by 84.88.76.49 -
Changed lines 7-10 from:

2008

Timeline:

to:

2008 Timeline

Changed lines 11-22 from:
  • 31 March 2008 - Submission deadline for the challenge

2007 (Archived)

The Web Spam Challenge 2007 was supported by the EU PASCAL Network of Excellence Challenge Program and had two tracks:

http://www.yr-bcn.es/webspam/graphs/pascal_logo.gif

  • January-May: Track I: focused on Information Retrieval and Machine Learning, jointly organized with the AIRWeb 2007 workshop.
  • June-September: Track II: focused on Machine Learning, jointly organized with the ECML/PKDD Workshop on Graph Labeling.
to:
  • 13 April 2008 - Submission deadline for the challenge
March 12, 2008, at 05:33 AM by 84.88.76.49 -
Added lines 5-6:

Where to start? Just download the WEBSPAM-UK2007 dataset and optionally some pre-computed feature vectors (available in Matlab and ARFF format).

March 06, 2008, at 03:11 PM by 84.88.76.49 -
Changed line 10 from:
to:
March 06, 2008, at 02:30 PM by 84.88.76.49 -
Changed line 10 from:
to:
March 06, 2008, at 02:30 PM by 84.88.76.49 -
Changed line 10 from:
  • mid-February 2008 (planned) - Pre-computed feature sets will be available
to:
January 18, 2008, at 11:18 AM by 84.88.76.49 -
Deleted lines 6-7:

A new dataset: WEBSPAM-UK2007, based on a crawl of .UK done on May 2007, is available.

Changed line 9 from:
  • 18 January 2008 - Dataset available
to:
  • 18 January 2008 - WEBSPAM-UK2007 dataset available
January 18, 2008, at 11:17 AM by 84.88.76.49 -
Added lines 9-14:

Timeline:

  • 18 January 2008 - Dataset available
  • mid-February 2008 (planned) - Pre-computed feature sets will be available
  • 31 March 2008 - Submission deadline for the challenge
January 16, 2008, at 09:00 AM by 84.88.76.49 -
Changed lines 5-10 from:

A new dataset: WEBSPAM-UK2007, based on a crawl of .UK done on May 2007, will be used.

Mailing list

If you are interested in participating in a future Web Spam Challenge, please subscribe to our mailing list.

to:

2008

A new dataset: WEBSPAM-UK2007, based on a crawl of .UK done on May 2007, is available.

Added lines 19-22:

Mailing list

If you are interested in participating in a future Web Spam Challenge, please subscribe to our mailing list.

January 16, 2008, at 09:00 AM by 84.88.76.49 -
Added lines 5-6:

A new dataset: WEBSPAM-UK2007, based on a crawl of .UK done on May 2007, will be used.

December 07, 2007, at 10:38 AM by 216.145.54.158 -
December 07, 2007, at 10:32 AM by 216.145.54.158 -
Added lines 5-8:

Mailing list

If you are interested in participating in a future Web Spam Challenge, please subscribe to our mailing list.

Deleted lines 18-20:

Mailing list

If you are interested in participating in the challenge, please subscribe to our mailing list.

December 07, 2007, at 10:32 AM by 216.145.54.158 -
Changed lines 3-4 from:

The goal of the Web Spam Challenge is to identify and compare Machine Learning (ML) methods for automatically labelling structured data represented as graphs. More precisely, we focus on the problem of labelling all nodes of a graph from a partial labelling of them. The application we study is Web Spam Detection, where we want to detect deliberate actions of deception aimed at the ranking functions used by search engines.

to:

The goal of the Web Spam Challenge series to identify and compare Machine Learning (ML) methods for automatically labelling structured data represented as graphs. More precisely, we focus on the problem of labelling all nodes of a graph from a partial labelling of them. The application we study is Web Spam Detection, where we want to detect deliberate actions of deception aimed at the ranking functions used by search engines.

2007 (Archived)

The Web Spam Challenge 2007 was supported by the EU PASCAL Network of Excellence Challenge Program and had two tracks:

Deleted lines 10-13:

The Web Spam Challenge is supported by the EU PASCAL Network of Excellence Challenge Program.

During 2007, the Web Spam Challenge will have two tracks:

December 07, 2007, at 10:11 AM by 216.145.54.158 -
Deleted lines 2-3:

The testing labels for part II are now available (small graph and large graph) - see Here

July 24, 2007, at 08:53 AM by 84.35.81.2 -
Changed lines 3-4 from:

The traning corpora for part II are now available (small graph and large graph) - see Here

to:

The testing labels for part II are now available (small graph and large graph) - see Here

June 11, 2007, at 07:43 AM by 132.227.204.229 -
Added lines 3-4:

The traning corpora for part II are now available (small graph and large graph) - see Here

May 31, 2007, at 09:49 AM by ChaTo -
Changed lines 11-14 from:
  • Track I: focused on Information Retrieval and Machine Learning, jointly organized with the AIRWeb 2007 workshop.
  • Track II: focused on Machine Learning, jointly organized with the ECML/PKDD Workshop on Graph Labeling.
to:
  • January-May: Track I: focused on Information Retrieval and Machine Learning, jointly organized with the AIRWeb 2007 workshop.
  • June-September: Track II: focused on Machine Learning, jointly organized with the ECML/PKDD Workshop on Graph Labeling.
May 02, 2007, at 09:47 AM by ChaTo -
Changed lines 11-14 from:

Track I: mainly about Information Retrieval and Machine Learning, jointly organized with the AIRWeb 2007 workshop.

Track II: mainly about Machine Learning, jointly organized with the ECML/PKDD Workshop on Graph Labeling.

to:
  • Track I: focused on Information Retrieval and Machine Learning, jointly organized with the AIRWeb 2007 workshop.
  • Track II: focused on Machine Learning, jointly organized with the ECML/PKDD Workshop on Graph Labeling.
May 02, 2007, at 09:47 AM by ChaTo -
May 02, 2007, at 09:47 AM by ChaTo -
Changed lines 11-14 from:

Track I: related to Information Retrieval and Machine Learning, jointly organized with the AIRWeb 2007 workshop.

Track II: related to Machine Learning, jointly organized with the ECML/PKDD Workshop on Graph Labeling.

to:

Track I: mainly about Information Retrieval and Machine Learning, jointly organized with the AIRWeb 2007 workshop.

Track II: mainly about Machine Learning, jointly organized with the ECML/PKDD Workshop on Graph Labeling.

May 02, 2007, at 09:46 AM by ChaTo -
Changed lines 11-14 from:

Track I: mostly directed to researchers on Information Retrieval and Machine Learning, jointly organized with the AIRWeb 2007 workshop.

Track II: mostly directed to researchers on Machine Learning, jointly organized with the ECML/PKDD Workshop on Graph Labeling.

to:

Track I: related to Information Retrieval and Machine Learning, jointly organized with the AIRWeb 2007 workshop.

Track II: related to Machine Learning, jointly organized with the ECML/PKDD Workshop on Graph Labeling.

May 02, 2007, at 09:46 AM by ChaTo -
Changed lines 13-18 from:

Track II: mostly directed to researchers on Machine Learning, being planned for the second half of 2007.

Timeline

  • 8 May 2007: Results of the evaluation announced at the AIRWeb workshop
to:

Track II: mostly directed to researchers on Machine Learning, jointly organized with the ECML/PKDD Workshop on Graph Labeling.

May 02, 2007, at 09:44 AM by ChaTo -
Changed line 16 from:
to:
Deleted lines 21-33:

History

April 30, 2007, at 04:17 PM by 82.67.192.190 -
April 10, 2007, at 03:41 AM by ChaTo -
April 10, 2007, at 03:41 AM by ChaTo -
Deleted lines 15-16:
Added lines 33-34:
March 29, 2007, at 10:06 AM by ChaTo -
Changed line 17 from:
to:
March 17, 2007, at 07:16 AM by ChaTo -
Changed lines 16-17 from:
to:
March 17, 2007, at 07:16 AM by ChaTo -
Changed line 17 from:
to:
February 06, 2007, at 07:50 AM by 132.227.204.229 -
Changed lines 34-35 from:
  • January 17, 2007: Challenge accepted by PASCAL Network
to:
  • January 17, 2007: Challenge accepted by PASCAL Network
  • February 2007: new features vectors available
January 17, 2007, at 11:20 AM by ChaTo -
Changed lines 7-8 from:

The Web Spam Challenge is supported by the EU PASCAL Network of Excellence Challenge Program'''.

to:

The Web Spam Challenge is supported by the EU PASCAL Network of Excellence Challenge Program.

January 17, 2007, at 11:20 AM by ChaTo -
Changed lines 7-8 from:

The Web Spam Challenge is supported by the EU PASCAL Network of Excellence Challenge Program.

to:

The Web Spam Challenge is supported by the EU PASCAL Network of Excellence Challenge Program'''.

January 17, 2007, at 11:19 AM by ChaTo -
Changed lines 5-6 from:

http://www.yr-bcn.es/webspam/graphs/pascal_logo.gif

to:

http://www.yr-bcn.es/webspam/graphs/pascal_logo.gif

January 17, 2007, at 11:19 AM by ChaTo -
Changed lines 5-6 from:

http://www.yr-bcn.es/webspam/graphs/pascal_logo.gif

to:

http://www.yr-bcn.es/webspam/graphs/pascal_logo.gif

January 17, 2007, at 11:19 AM by ChaTo -
Changed lines 11-14 from:
  • Track I: mostly directed to researchers on Information Retrieval and Machine Learning, jointly organized with the AIRWeb 2007 workshop.
  • Track II: mostly directed to researchers on Machine Learning, being planned for the second half of 2007.
to:

Track I: mostly directed to researchers on Information Retrieval and Machine Learning, jointly organized with the AIRWeb 2007 workshop.

Track II: mostly directed to researchers on Machine Learning, being planned for the second half of 2007.

January 17, 2007, at 11:19 AM by ChaTo -
Changed lines 11-14 from:

Track I: mostly directed to researchers on Information Retrieval and Machine Learning, jointly organized with the AIRWeb 2007 workshop.

Track II: mostly directed to researchers on Machine Learning, being planned for the second half of 2007.

to:
  • Track I: mostly directed to researchers on Information Retrieval and Machine Learning, jointly organized with the AIRWeb 2007 workshop.
  • Track II: mostly directed to researchers on Machine Learning, being planned for the second half of 2007.
January 17, 2007, at 11:19 AM by ChaTo -
Changed lines 5-7 from:

http://www.yr-bcn.es/webspam/graphs/pascal_logo.gif

to:

http://www.yr-bcn.es/webspam/graphs/pascal_logo.gif

Changed lines 8-10 from:
to:
January 17, 2007, at 11:18 AM by ChaTo -
Added lines 1-4:

Welcome

The goal of the Web Spam Challenge is to identify and compare Machine Learning (ML) methods for automatically labelling structured data represented as graphs. More precisely, we focus on the problem of labelling all nodes of a graph from a partial labelling of them. The application we study is Web Spam Detection, where we want to detect deliberate actions of deception aimed at the ranking functions used by search engines.

Changed lines 11-14 from:

Welcome

The goal of the Web Spam Challenge is to identify and compare Machine Learning (ML) methods for automatically labelling structured data represented as graphs. More precisely, we focus on the problem of labelling all nodes of a graph from a partial labelling of them. The application we study is Web Spam Detection, where we want to detect deliberate actions of deception aimed at the ranking functions used by search engines.

to:
January 17, 2007, at 11:04 AM by ChaTo -
Changed lines 7-8 from:

Web Spam Challenge

to:

Welcome

January 17, 2007, at 11:04 AM by ChaTo -
Changed lines 7-8 from:

The Web Spam Challenge

to:

Web Spam Challenge

January 17, 2007, at 11:04 AM by ChaTo -
Changed lines 7-8 from:

Welcome

to:

The Web Spam Challenge

January 17, 2007, at 10:04 AM by ChaTo -
Changed line 1 from:
to:
January 17, 2007, at 10:03 AM by ChaTo -
Changed line 1 from:
to:
January 17, 2007, at 10:03 AM by ChaTo -
Changed line 1 from:
to:
Changed line 4 from:

The Web Spam Challenge is supported by the EU PASCAL Network of Excellence challenge program.

to:

The Web Spam Challenge is supported by the EU PASCAL Network of Excellence Challenge Program.

January 17, 2007, at 10:03 AM by ChaTo -
Changed line 4 from:

The Web Spam Challenge is supported by the EU PASCAL Network of Excellence Challenge program.

to:

The Web Spam Challenge is supported by the EU PASCAL Network of Excellence challenge program.

January 17, 2007, at 10:02 AM by ChaTo -
Deleted lines 16-25:

History

Added lines 28-36:

History

January 17, 2007, at 09:59 AM by ChaTo -
Changed lines 2-3 from:

http://www.yr-bcn.es/webspam/graphs/pascal_logo.gif

to:

http://www.yr-bcn.es/webspam/graphs/pascal_logo.gif

January 17, 2007, at 09:59 AM by ChaTo -
Added line 3:
January 17, 2007, at 09:58 AM by ChaTo -
Added line 2:

http://www.yr-bcn.es/webspam/graphs/pascal_logo.gif

Deleted lines 34-35:

http://www.yr-bcn.es/webspam/graphs/pascal_logo.gif

January 17, 2007, at 09:58 AM by ChaTo -
Added lines 34-35:

http://www.yr-bcn.es/webspam/graphs/pascal_logo.gif

Deleted line 37:

http://www.yr-bcn.es/webspam/graphs/pascal_logo.gif | Part of the EU PASCAL Network of Excellence Challenge program.

January 17, 2007, at 09:58 AM by ChaTo -
Changed line 36 from:
to:

http://www.yr-bcn.es/webspam/graphs/pascal_logo.gif | Part of the EU PASCAL Network of Excellence Challenge program.

January 17, 2007, at 09:55 AM by ChaTo -
Changed line 2 from:

The Web Spam Challenge is a comparison of web spam detection systems supported by the EU PASCAL Network of Excellence Challenge program.

to:

The Web Spam Challenge is supported by the EU PASCAL Network of Excellence Challenge program.

January 17, 2007, at 09:55 AM by ChaTo -
Changed line 2 from:

The Web Spam Challenge is supported by the EU PASCAL Network of Excellence Challenge program.

to:

The Web Spam Challenge is a comparison of web spam detection systems supported by the EU PASCAL Network of Excellence Challenge program.

January 17, 2007, at 09:54 AM by ChaTo -
Changed line 2 from:

The Web Spam Challenge is a challenge for testing web spam detection systems, and is supported by the EU PASCAL Network of Excellence.

to:

The Web Spam Challenge is supported by the EU PASCAL Network of Excellence Challenge program.

January 17, 2007, at 09:54 AM by ChaTo -
January 17, 2007, at 09:53 AM by ChaTo -
Changed line 2 from:

The Web Spam Challenge is a benchmark for testing web spam detection systems, supported by the EU PASCAL Network of Excellence.

to:

The Web Spam Challenge is a challenge for testing web spam detection systems, and is supported by the EU PASCAL Network of Excellence.

January 17, 2007, at 09:42 AM by ChaTo -
Changed line 2 from:

The Web Spam Challenge is a challenge for testing web spam detection systems, supported by the EU PASCAL Network of Excellence.

to:

The Web Spam Challenge is a benchmark for testing web spam detection systems, supported by the EU PASCAL Network of Excellence.

January 17, 2007, at 09:41 AM by ChaTo -
Changed line 2 from:

The Web Spam Challenge is a competition for testing web spam detection systems, supported by the EU PASCAL Network of Excellence.

to:

The Web Spam Challenge is a challenge for testing web spam detection systems, supported by the EU PASCAL Network of Excellence.

January 17, 2007, at 09:40 AM by ChaTo -
Changed lines 1-2 from:

The Web Spam Challenge is a competition for testing web spam detection systems, supported by the EU PASCAL Network of Excellence.

to:

The Web Spam Challenge is a competition for testing web spam detection systems, supported by the EU PASCAL Network of Excellence.

January 17, 2007, at 09:40 AM by ChaTo -
Changed lines 1-2 from:

% The Web Spam Challenge is a competition for testing web spam detection systems, supported by the EU PASCAL Network of Excellence.

to:

The Web Spam Challenge is a competition for testing web spam detection systems, supported by the EU PASCAL Network of Excellence.

January 17, 2007, at 09:40 AM by ChaTo -
Changed lines 1-2 from:

The Web Spam Challenge is a competition for testing web spam detection systems, supported by the EU PASCAL Network of Excellence.

to:

% The Web Spam Challenge is a competition for testing web spam detection systems, supported by the EU PASCAL Network of Excellence.

January 17, 2007, at 09:39 AM by ChaTo -
Changed lines 1-2 from:

The Web Spam Challenge is a competition for testing web spam detection systems, supported by the EU PASCAL Network of Excellence.

to:

The Web Spam Challenge is a competition for testing web spam detection systems, supported by the EU PASCAL Network of Excellence.

January 17, 2007, at 09:39 AM by ChaTo -
Changed lines 1-2 from:

%bold color=FF0000% The Web Spam Challenge is a competition for testing web spam detection systems, supported by the EU PASCAL Network of Excellence.

to:

The Web Spam Challenge is a competition for testing web spam detection systems, supported by the EU PASCAL Network of Excellence.

January 17, 2007, at 09:39 AM by ChaTo -
Changed lines 1-2 from:

%color=FF0000% The Web Spam Challenge is a competition for testing web spam detection systems, supported by the EU PASCAL Network of Excellence.

to:

%bold color=FF0000% The Web Spam Challenge is a competition for testing web spam detection systems, supported by the EU PASCAL Network of Excellence.

January 17, 2007, at 09:39 AM by ChaTo -
Changed lines 1-2 from:

%color=#FF0000% The Web Spam Challenge is a competition for testing web spam detection systems, supported by the EU PASCAL Network of Excellence.

to:

%color=FF0000% The Web Spam Challenge is a competition for testing web spam detection systems, supported by the EU PASCAL Network of Excellence.

January 17, 2007, at 09:39 AM by ChaTo -
Changed lines 1-2 from:

%black color=#FF0000% The Web Spam Challenge is a competition for testing web spam detection systems, supported by the EU PASCAL Network of Excellence.

to:

%color=#FF0000% The Web Spam Challenge is a competition for testing web spam detection systems, supported by the EU PASCAL Network of Excellence.

January 17, 2007, at 09:39 AM by ChaTo -
Changed lines 1-2 from:

The Web Spam Challenge is a competition for testing web spam detection systems, supported by the EU PASCAL Network of Excellence.

to:

%black color=#FF0000% The Web Spam Challenge is a competition for testing web spam detection systems, supported by the EU PASCAL Network of Excellence.

January 17, 2007, at 09:37 AM by ChaTo -
Changed lines 21-22 from:
  • January 17, 2007: Challenge accepted by PASCAL Network
to:
  • January 17, 2007: Challenge accepted by PASCAL Network
January 17, 2007, at 09:36 AM by ChaTo -
Changed lines 1-2 from:

Web Spam Challenge -- supported by the PASCAL Network

to:

The Web Spam Challenge is a competition for testing web spam detection systems, supported by the EU PASCAL Network of Excellence.

January 17, 2007, at 09:36 AM by ChaTo -
Changed lines 1-3 from:

Web Spam Challenge

Supported by the PASCAL Network

to:

Web Spam Challenge -- supported by the PASCAL Network

January 17, 2007, at 09:35 AM by ChaTo -
Changed lines 1-2 from:

Web Spam Challenge — supported by PASCAL Network

to:

Web Spam Challenge

Supported by the PASCAL Network

January 17, 2007, at 09:35 AM by ChaTo -
Changed lines 1-2 from:

Web Spam Challenge — supported by the PASCAL Network

to:

Web Spam Challenge — supported by PASCAL Network

January 17, 2007, at 09:35 AM by ChaTo -
Changed lines 1-2 from:

Web Spam Challenge - supported by the PASCAL Network

to:

Web Spam Challenge — supported by the PASCAL Network

January 17, 2007, at 09:24 AM by ChaTo -
Changed line 19 from:
to:
Added line 25:
January 17, 2007, at 09:23 AM by ChaTo -
Changed lines 20-21 from:
  • January 17, 2006: Challenge accepted by PASCAL Network
to:
Deleted line 24:
January 17, 2007, at 09:23 AM by ChaTo -
January 17, 2007, at 09:22 AM by ChaTo -
Changed lines 20-21 from:
  • January 2006: Challenge accepted by PASCAL Network
to:
  • January 17, 2006: Challenge accepted by PASCAL Network
January 17, 2007, at 09:03 AM by 82.67.192.190 -
Changed line 20 from:
  • January 2006: Challenge accpeted by PASCAL Network
to:
  • January 2006: Challenge accepted by PASCAL Network
January 17, 2007, at 09:03 AM by 82.67.192.190 -
Changed line 20 from:
to:
  • January 2006: Challenge accpeted by PASCAL Network
January 17, 2007, at 08:56 AM by 82.67.192.190 -
Changed lines 1-2 from:

Web Spam Challenge

to:

Web Spam Challenge - supported by the PASCAL Network

January 11, 2007, at 09:49 AM by ChaTo -
Changed lines 13-14 from:

Timeline/News

to:

History

Added lines 20-22:

Timeline

January 11, 2007, at 09:49 AM by ChaTo -
Changed line 16 from:
  • November 2006: The AIRWeb Workshop has been accepted at WWW'07. The first part of the Web Spam Challenge will be co-organized with AIRWeb.
to:
  • November 2006: The AIRWeb Workshop was accepted at WWW'07.
Deleted line 18:
Deleted line 19:
January 11, 2007, at 09:48 AM by ChaTo -
Changed lines 13-17 from:
to:

Timeline/News

Changed lines 31-41 from:

News

January 2, 2007: A set of link-based pre-computed features is available.

December 22, 2006: host graph is available.

December 2006: The website is open.

November 2006: The AIRWeb Workshop has been accepted at WWW'07. The first part of the Web Spam Challenge will be co-organized with AIRWeb.

September 2006: The challenge has been submitted to the PASCAL Network

to:
January 11, 2007, at 09:47 AM by ChaTo -
Added lines 13-21:

Timeline

January 03, 2007, at 10:42 AM by ChaTo -
Changed lines 19-22 from:

2/January/2007: A set of link-based pre-computed features is available.

22/December/2006: host graph is available.

to:

January 2, 2007: A set of link-based pre-computed features is available.

December 22, 2006: host graph is available.

January 03, 2007, at 10:42 AM by ChaTo -
Added line 20:
January 03, 2007, at 10:42 AM by ChaTo -
Changed lines 19-20 from:

22nd of December 2006: host graph is available.

to:

2/January/2007: A set of link-based pre-computed features is available. 22/December/2006: host graph is available.

December 27, 2006, at 05:56 AM by ChaTo -
Changed lines 19-20 from:

22nd of December 2006: Host graph, host index and host labels are downloadable

to:

22nd of December 2006: host graph is available.

December 22, 2006, at 11:51 AM by 82.67.192.190 -
Changed lines 19-22 from:

22nd of Dececember 2006: Host graph, host index and host labels are downloadable

Dececember 2006: The website is open.

to:

22nd of December 2006: Host graph, host index and host labels are downloadable

December 2006: The website is open.

December 22, 2006, at 11:51 AM by 82.67.192.190 -
Added lines 19-20:

22nd of Dececember 2006: Host graph, host index and host labels are downloadable

December 12, 2006, at 09:35 AM by 84.88.76.30 -
Changed lines 15-16 from:

If you are interested in participating in the challenge, please subscribe to our mailing list.

to:

If you are interested in participating in the challenge, please subscribe to our mailing list.

December 10, 2006, at 08:18 PM by 207.172.52.93 -
Changed lines 21-22 from:

November 2006: The AirWeb Workshop has been accepted at WWW'07. The first part of the Web Spam Challenge will be co-organized with AIRWeb.

to:

November 2006: The AIRWeb Workshop has been accepted at WWW'07. The first part of the Web Spam Challenge will be co-organized with AIRWeb.

December 06, 2006, at 08:04 AM by 84.88.76.30 -
Changed lines 15-16 from:

Please subscribe to our mailing list.

to:

If you are interested in participating in the challenge, please subscribe to our mailing list.

December 06, 2006, at 08:04 AM by 84.88.76.30 -
Added lines 13-16:

Mailing list

Please subscribe to our mailing list.

December 06, 2006, at 08:03 AM by 84.88.76.30 -
Changed lines 15-19 from:

6 Dec 2006: The website is open.

Nov 2006: The AirWeb Workshop has been accepted at WWW'07. The first part of the Web Spam Challenge will be co-organized with AIRWeb.

Sept 2006 The challenge has been submitted to the PASCAL Network

to:

Dececember 2006: The website is open.

November 2006: The AirWeb Workshop has been accepted at WWW'07. The first part of the Web Spam Challenge will be co-organized with AIRWeb.

September 2006: The challenge has been submitted to the PASCAL Network

December 06, 2006, at 07:47 AM by 84.88.76.30 -
Changed lines 5-8 from:

The goal of the Web Spam Challenge is to identify and compare Machine Learning (ML) methods for automatically labelling structured data represented as graphs. More precisely, we focus on the problem of labelling all nodes of a graph from a partial labelling of them.

We focus here on the specific application of Web Spam Labelling where we want to detect deliberate actions of deception aimed at the ranking functions used by search engines.

to:

The goal of the Web Spam Challenge is to identify and compare Machine Learning (ML) methods for automatically labelling structured data represented as graphs. More precisely, we focus on the problem of labelling all nodes of a graph from a partial labelling of them. The application we study is Web Spam Detection, where we want to detect deliberate actions of deception aimed at the ranking functions used by search engines.

December 06, 2006, at 07:20 AM by 84.88.76.30 -
Deleted lines 4-5:

Learning to Label a Graph: Application to Web Spam Detection

December 06, 2006, at 06:50 AM by 84.88.76.30 -
December 06, 2006, at 06:49 AM by 84.88.76.30 -
Changed lines 11-12 from:

The Web Spam Challenge will be carried in two phases:

to:

During 2007, the Web Spam Challenge will have two tracks:

December 06, 2006, at 06:49 AM by 84.88.76.30 -
Changed lines 15-16 from:

'Track II: mostly directed to researchers on Machine Learning, being planned for the second half of 2007.

to:

Track II: mostly directed to researchers on Machine Learning, being planned for the second half of 2007.

December 06, 2006, at 06:49 AM by 84.88.76.30 -
Changed lines 15-16 from:

Phase 2: mostly directed to researchers on Machine Learning, being planned for the second half of 2007.

to:

'Track II: mostly directed to researchers on Machine Learning, being planned for the second half of 2007.

December 06, 2006, at 06:48 AM by 84.88.76.30 -
Changed lines 13-14 from:

Phase 1: mostly directed to researchers on Information Retrieval and Machine Learning, jointly organized with the AIRWeb 2007 workshop.

to:

Track I: mostly directed to researchers on Information Retrieval and Machine Learning, jointly organized with the AIRWeb 2007 workshop.

December 06, 2006, at 06:43 AM by 84.88.76.30 -
Changed lines 13-14 from:

Phase 1: mostly directed to researchers on Information Retrieval and Machine Learning, jointly organized with the AIRWeb 2007 workshop.

to:

Phase 1: mostly directed to researchers on Information Retrieval and Machine Learning, jointly organized with the AIRWeb 2007 workshop.

December 06, 2006, at 06:39 AM by 84.88.76.30 -
Changed lines 3-4 from:

Learning to Label a Graph: Application to Web Spam Detection

to:

Welcome

Learning to Label a Graph: Application to Web Spam Detection

Changed lines 17-18 from:

News

to:

News

December 06, 2006, at 06:39 AM by 84.88.76.30 -
Changed lines 15-16 from:

News

to:

News

December 06, 2006, at 06:38 AM by 84.88.76.30 -
Changed lines 15-16 from:

News

to:

News

December 06, 2006, at 06:12 AM by 84.88.76.30 -
Changed lines 7-8 from:

We focus here on the specific application of Web Spam Labelling where we want to detect deliberate actions of deception aimed at the ranking functions used by search engines.

to:

We focus here on the specific application of Web Spam Labelling where we want to detect deliberate actions of deception aimed at the ranking functions used by search engines.

The Web Spam Challenge will be carried in two phases:

Phase 1: mostly directed to researchers on Information Retrieval and Machine Learning, jointly organized with the AIRWeb 2007 workshop.

Phase 2: mostly directed to researchers on Machine Learning, being planned for the second half of 2007.

December 06, 2006, at 06:05 AM by 84.88.76.30 -
Changed lines 11-13 from:
  • 06 Dec 2006: The website is open.
  • Nov 2006: The AirWeb Workshop has been accepted at WWW'07. The first part of the Web Spam Challenge will be co-organized with AIRWeb.
  • Sept 2006 The challenge has been submitted to the PASCAL Network
to:

6 Dec 2006: The website is open.

Nov 2006: The AirWeb Workshop has been accepted at WWW'07. The first part of the Web Spam Challenge will be co-organized with AIRWeb.

Sept 2006 The challenge has been submitted to the PASCAL Network

December 06, 2006, at 06:04 AM by 84.88.76.30 -
Changed lines 1-4 from:

Learning to Label a Graph

Application to Web Spam Detection

to:

Web Spam Challenge

Learning to Label a Graph: Application to Web Spam Detection

December 06, 2006, at 06:04 AM by 84.88.76.30 -
Changed lines 1-4 from:

Learning to Label a Graph

Application to Web Spam Detection

to:

Learning to Label a Graph

Application to Web Spam Detection

December 06, 2006, at 06:04 AM by 84.88.76.30 -
Changed lines 9-13 from:

News

06 Dec 2006
The website is open.
Nov 2006
The AirWeb Workshop has been accepted at WWW'07. The first part of the Web Spam Challenge will be co-organized with AIRWeb.
Sept 2006
The challenge has been submitted to the PASCAL Network
to:

News

  • 06 Dec 2006: The website is open.
  • Nov 2006: The AirWeb Workshop has been accepted at WWW'07. The first part of the Web Spam Challenge will be co-organized with AIRWeb.
  • Sept 2006 The challenge has been submitted to the PASCAL Network
December 06, 2006, at 06:03 AM by 84.88.76.30 -
Changed lines 1-5 from:

Learning to Label a Graph: Application to Web Spam Detection

The goal of the challenge is to identify and compare Machine Learning (ML) methods for automatically labelling structured data represented as graphs. More precisely, we focus on the problem of labelling all nodes of a graph from a partial labelling of this graph as described in figure 1.

to:

Learning to Label a Graph

Application to Web Spam Detection

The goal of the Web Spam Challenge is to identify and compare Machine Learning (ML) methods for automatically labelling structured data represented as graphs. More precisely, we focus on the problem of labelling all nodes of a graph from a partial labelling of them.

Deleted lines 8-10:
http://www-connex.lip6.fr/~denoyer/ws/graph.png
Figure 1 : An example of graph to label
Changed lines 11-13 from:
06 Dec 2006
The website is opened
Nov 2006
The AirWeb Workshop has been accepted at WWW'07. The challenge will be strongly connected to this workshop
Sept 2006
The challenge has been submitted at the PASCAL Network
to:
06 Dec 2006
The website is open.
Nov 2006
The AirWeb Workshop has been accepted at WWW'07. The first part of the Web Spam Challenge will be co-organized with AIRWeb.
Sept 2006
The challenge has been submitted to the PASCAL Network
December 06, 2006, at 03:31 AM by 82.67.192.190 -
Changed lines 11-15 from:
to:

News

06 Dec 2006
The website is opened
Nov 2006
The AirWeb Workshop has been accepted at WWW'07. The challenge will be strongly connected to this workshop
Sept 2006
The challenge has been submitted at the PASCAL Network
December 06, 2006, at 03:07 AM by 82.67.192.190 -
Changed lines 4-5 from:

The goal of the challenge is to identify and compare Machine Learning (ML) methods for automatically labelling structured data represented as graphs. More precisely, we focus on the problem of labelling all nodes of a graph from a partial labelling of this graph.

to:

The goal of the challenge is to identify and compare Machine Learning (ML) methods for automatically labelling structured data represented as graphs. More precisely, we focus on the problem of labelling all nodes of a graph from a partial labelling of this graph as described in figure 1.

Changed lines 9-10 from:
to:
Figure 1 : An example of graph to label
December 06, 2006, at 03:06 AM by 82.67.192.190 -
Changed lines 8-9 from:

http://www-connex.lip6.fr/~denoyer/ws/graph.png

to:
http://www-connex.lip6.fr/~denoyer/ws/graph.png
December 06, 2006, at 03:04 AM by 82.67.192.190 -
Changed lines 8-9 from:

http://www-connex.lip6.fr/~denoyer/ws/graph.png

to:

http://www-connex.lip6.fr/~denoyer/ws/graph.png

December 06, 2006, at 03:04 AM by 82.67.192.190 -
Changed lines 3-5 from:

Introduction

The goal of the challenge is to identify and compare Machine Learning (ML) methods for automatically labelling structured data represented as graphs. More precisely, we focus on the problem of labelling all nodes of a graph from a partial labelling of this graph as depicted in Figure 1. Many practical applications are instances of this generic machine learning problem. We focus here on the specific application of Web Spam Labelling where we want to detect deliberate actions of deception aimed at the ranking functions used by search engines. The nodes of the graph can be either individual Web pages or groups of pages, such as entire domains. The labels are either categories (spam or non-spam) for Web Spam Detection or numbers reflecting the “spamicity” of a page for Web Spam Demotion. In the first case, the system must predict the binary labels while in the second case the system must provide an estimator of the actual “spamicity” of each object. The automatic labeling systems may use all the aspects of the corpus as sources of evidence. One source is the content of the Web pages themselves, this is, their text, images and HTML markup. Another source is the Web graph, this is, a conceptualization of the Web as a graph in which the nodes are pages and the edges are the hyperlinks between them.

to:

The goal of the challenge is to identify and compare Machine Learning (ML) methods for automatically labelling structured data represented as graphs. More precisely, we focus on the problem of labelling all nodes of a graph from a partial labelling of this graph.

We focus here on the specific application of Web Spam Labelling where we want to detect deliberate actions of deception aimed at the ranking functions used by search engines.

http://www-connex.lip6.fr/~denoyer/ws/graph.png

December 05, 2006, at 10:30 AM by 82.67.192.190 -
Changed line 5 from:
 The goal of the challenge is to identify and compare Machine Learning (ML) methods for automatically labelling structured data represented as graphs.  More precisely, we focus on the problem of labelling all nodes of a graph from a partial labelling of this graph as depicted in Figure 1. Many practical applications are instances of this generic machine learning problem. We focus here on the specific application of Web Spam Labelling where we want to detect deliberate actions of deception aimed at the ranking functions used by search engines. The nodes of the graph can be either individual Web pages or groups of pages, such as entire domains. The labels are either categories (spam or non-spam) for Web Spam Detection or numbers reflecting the “spamicity” of a page for Web Spam Demotion. In the first case, the system must predict the binary labels while in the second case the system must provide an estimator of the actual “spamicity” of each object. The automatic labeling systems may use all the aspects of the corpus as sources of evidence. One source is the content of the Web pages themselves, this is, their text, images and HTML markup. Another source is the Web graph, this is, a conceptualization of the Web as a graph in which the nodes are pages and the edges are the hyperlinks between them.
to:

The goal of the challenge is to identify and compare Machine Learning (ML) methods for automatically labelling structured data represented as graphs. More precisely, we focus on the problem of labelling all nodes of a graph from a partial labelling of this graph as depicted in Figure 1. Many practical applications are instances of this generic machine learning problem. We focus here on the specific application of Web Spam Labelling where we want to detect deliberate actions of deception aimed at the ranking functions used by search engines. The nodes of the graph can be either individual Web pages or groups of pages, such as entire domains. The labels are either categories (spam or non-spam) for Web Spam Detection or numbers reflecting the “spamicity” of a page for Web Spam Demotion. In the first case, the system must predict the binary labels while in the second case the system must provide an estimator of the actual “spamicity” of each object. The automatic labeling systems may use all the aspects of the corpus as sources of evidence. One source is the content of the Web pages themselves, this is, their text, images and HTML markup. Another source is the Web graph, this is, a conceptualization of the Web as a graph in which the nodes are pages and the edges are the hyperlinks between them.

December 05, 2006, at 10:30 AM by 82.67.192.190 -
Changed lines 5-11 from:
 The goal of the challenge is to identify and compare Machine Learning (ML)
 methods for automatically labelling structured data represented as graphs.
 More precisely, we focus on the problem of labelling all nodes of a graph 

from a partial labelling of this graph as depicted in Figure 1. Many practical

 applications are instances of this generic machine learning problem. We focus 

here on the specific application of Web Spam Labelling where we want to detect deliberate actions of deception aimed at the ranking functions used by search engines. The nodes of the graph can be either individual Web pages or groups of pages, such as entire domains. The labels are either categories (spam or non-spam) for Web Spam Detection or numbers reflecting the “spamicity” of a page for Web Spam Demotion. In the first case, the system must predict the binary labels while in the second case the system must provide an estimator of the actual “spamicity” of each object. The automatic labeling systems may use all the aspects of the corpus as sources of evidence. One source is the content of the Web pages themselves, this is, their text, images and HTML markup. Another source is the Web graph, this is, a conceptualization of the Web as a graph in which the nodes are pages and the edges are the hyperlinks between them.

to:
 The goal of the challenge is to identify and compare Machine Learning (ML) methods for automatically labelling structured data represented as graphs.  More precisely, we focus on the problem of labelling all nodes of a graph from a partial labelling of this graph as depicted in Figure 1. Many practical applications are instances of this generic machine learning problem. We focus here on the specific application of Web Spam Labelling where we want to detect deliberate actions of deception aimed at the ranking functions used by search engines. The nodes of the graph can be either individual Web pages or groups of pages, such as entire domains. The labels are either categories (spam or non-spam) for Web Spam Detection or numbers reflecting the “spamicity” of a page for Web Spam Demotion. In the first case, the system must predict the binary labels while in the second case the system must provide an estimator of the actual “spamicity” of each object. The automatic labeling systems may use all the aspects of the corpus as sources of evidence. One source is the content of the Web pages themselves, this is, their text, images and HTML markup. Another source is the Web graph, this is, a conceptualization of the Web as a graph in which the nodes are pages and the edges are the hyperlinks between them.
December 05, 2006, at 10:29 AM by 82.67.192.190 -
Changed lines 5-11 from:
 The goal of the challenge is to identify and compare Machine Learning (ML) methods for automatically labelling structured data represented as graphs. More precisely, we focus on the problem of labelling all nodes of a graph from a partial labelling of this graph as depicted in Figure 1. Many practical applications are instances of this generic machine learning problem. We focus here on the specific application of Web Spam Labelling where we want to detect deliberate actions of deception aimed at the ranking functions used by search engines. The nodes of the graph can be either individual Web pages or groups of pages, such as entire domains. The labels are either categories (spam or non-spam) for Web Spam Detection or numbers reflecting the “spamicity” of a page for Web Spam Demotion. In the first case, the system must predict the binary labels while in the second case the system must provide an estimator of the actual “spamicity” of each object. The automatic labeling systems may use all the aspects of the corpus as sources of evidence. One source is the content of the Web pages themselves, this is, their text, images and HTML markup. Another source is the Web graph, this is, a conceptualization of the Web as a graph in which the nodes are pages and the edges are the hyperlinks between them.
to:
 The goal of the challenge is to identify and compare Machine Learning (ML)
 methods for automatically labelling structured data represented as graphs.
 More precisely, we focus on the problem of labelling all nodes of a graph 

from a partial labelling of this graph as depicted in Figure 1. Many practical

 applications are instances of this generic machine learning problem. We focus 

here on the specific application of Web Spam Labelling where we want to detect deliberate actions of deception aimed at the ranking functions used by search engines. The nodes of the graph can be either individual Web pages or groups of pages, such as entire domains. The labels are either categories (spam or non-spam) for Web Spam Detection or numbers reflecting the “spamicity” of a page for Web Spam Demotion. In the first case, the system must predict the binary labels while in the second case the system must provide an estimator of the actual “spamicity” of each object. The automatic labeling systems may use all the aspects of the corpus as sources of evidence. One source is the content of the Web pages themselves, this is, their text, images and HTML markup. Another source is the Web graph, this is, a conceptualization of the Web as a graph in which the nodes are pages and the edges are the hyperlinks between them.

December 05, 2006, at 10:29 AM by 82.67.192.190 -
Changed lines 2-5 from:

coucou

to:

Introduction

 The goal of the challenge is to identify and compare Machine Learning (ML) methods for automatically labelling structured data represented as graphs. More precisely, we focus on the problem of labelling all nodes of a graph from a partial labelling of this graph as depicted in Figure 1. Many practical applications are instances of this generic machine learning problem. We focus here on the specific application of Web Spam Labelling where we want to detect deliberate actions of deception aimed at the ranking functions used by search engines. The nodes of the graph can be either individual Web pages or groups of pages, such as entire domains. The labels are either categories (spam or non-spam) for Web Spam Detection or numbers reflecting the “spamicity” of a page for Web Spam Demotion. In the first case, the system must predict the binary labels while in the second case the system must provide an estimator of the actual “spamicity” of each object. The automatic labeling systems may use all the aspects of the corpus as sources of evidence. One source is the content of the Web pages themselves, this is, their text, images and HTML markup. Another source is the Web graph, this is, a conceptualization of the Web as a graph in which the nodes are pages and the edges are the hyperlinks between them.
December 05, 2006, at 10:29 AM by 82.67.192.190 -
Added line 2:

coucou

December 05, 2006, at 10:26 AM by 82.67.192.190 -
Changed lines 1-12 from:

Welcome to PmWiki! la la la A local copy of PmWiki's documentation has been installed along with the software, and is available via the documentation index.

To continue setting up PmWiki, see initial setup tasks.

The basic editing page describes how to create pages in PmWiki. You can practice editing in the wiki sandbox.

More information about PmWiki is available from http://www.pmwiki.org .

to:

Learning to Label a Graph: Application to Web Spam Detection

December 05, 2006, at 10:18 AM by 82.67.192.190 -
Changed line 2 from:
to:

la la la