nullification test collections for web spam and seo
play

Nullification test collections for Web spam and SEO Timothy Jones - PowerPoint PPT Presentation

Aug 03, 2023 •243 likes •390 views

Nullification test collections for Web spam and SEO Timothy Jones (ANU) David Hawking (Funnelback) Ramesh Sankaranarayana (ANU) Nick Craswel (Microsoft Research) Detection Detecting problem content Nullification Preventing problem content

  1. Nullification test collections for Web spam and SEO Timothy Jones (ANU) David Hawking (Funnelback) Ramesh Sankaranarayana (ANU) Nick Craswel (Microsoft Research)

  2. Detection Detecting problem content Nullification Preventing problem content from negatively affecting search results

  3. The UK-2006 and UK-2007 collections • Limited to only pages from the .uk TLD • 80M pages (UK-2006) 100M pages (UK-2007) • 10k hosts (UK-2006) 100k hosts (UK-2007) • Labelled partially at the host level o Spam o Non-spam o Borderline o Cannot classify

  4. Evaluating nullification with labels Can test a new ranking by using spam/non spam labels "Are spam pages demoted by this nullification?" However, not spam is not the same as relevant Need to check that relevant pages are not also demoted

  5. Evaluating nullification with users Preselected information need User picks information need • Need to have good • Collection must be current • Collection needs to be answers in the collection • Collection needs to be relevant to users relevant to users

  6. The UK-2006 and UK-2007 collections Well support testing spam detection Because of the domain limitation, collection only relevant to UK users Additionally, the structure may not be representative Companion sites • Popular queries • Graph statistics •

  7. Companion sites in UK-2007 There are many companion sites missing 1click-insurance.co.uk -> 1click-insurance.com 1click2keys.co.uk -> 1click2keys.com 1click2keys-overseas.co.uk -> 1click2keys-overseas.com 1click2keysoverseas.co.uk -> 1click2keysoverseas.com ... 3com.co.uk -> 3com.ch,3com.com,3com.cz,3com.de, 3com.fr, 3com.nl,3com.se ... abbott.co.uk -> abbott.com,abbott.de,abbott.dk, abbott.es, abbott.gr,abbott.ie,abbott.it, abbott.no There are 68,000 examples of this in UK-2007

  8. Companion sites in UK-2007 We used a structural heuristic to detect Single Entity Controlled Domains (SECDs) eg news.bbc.co.uk and www.bbc.co.uk More than 2.4 million non-.uk SECDs referenced Very high link counts from some .uk domains to non-.uk 11,000 non .uk domains linked to over 100,000 times from a single SECD each

  9. Popular queries in UK-2007 Popular queries are highly targeted by spammers Submitted the top 10 queries from the UK (using Google Zeitgeist's year end 2008 list) To two well known search engines Only 27.5% of the results were from .uk URLs Only 17% of the results were present in the collection

  10. Graph structure Power law exponent 1.7 when Frequency expecting of pages 2.1 In-degree

  11. The ideal collection • Page content • Link graph • Query and click data • Large • Recent • Spam/non spam labels • Sample queries known to be targetted by spam o With partial relevance judgements or o Information need statements for users

  12. Stanford WebBase Monthly crawls of 61 to 81 million pages Most recent crawl has 36,000 hosts 35% of the popular query results are present No spam labels May not contain much spam

  13. web09-bst Also known as ClueWeb09, to be used at TREC this year Contains 1 billion pages Intentionally designed to contain multi-lingual content A 50 million page English subset is available No spam labels Queries and click data are likely to be available for the TREC web track http://boston.lti.cs.cmu.edu/Data/web09-bst/

  14. Summary Evaluation of spam nullification is important, and can't be done with only spam/non-spam labels Need queries and relevance judgements Domain limited collections are unlikely to represent the web For evaluation of nullification, queries, relevance judgements, and a representative web crawl are essential The web09-bst / ClueWeb09 collection is likely to provide these

Download Presentation
Download Policy: The content available on the website is offered to you 'AS IS' for your personal information and use only. It cannot be commercialized, licensed, or distributed on other websites without prior consent from the author. To download a presentation, simply click this link. If you encounter any difficulties during the download process, it's possible that the publisher has removed the file from their server.

Recommend

Spam, Spam, Spam Why is spam interesting? Everyone can observe spam. Spam / Anti-spam is a

Spam, Spam, Spam Why is spam interesting? Everyone can observe spam. Spam / Anti-spam is a

Spam, Spam, Spam Why is spam interesting? Everyone can observe spam. Spam / Anti-spam is a highly evolved form of information warfare. Fascinating socioeconomic study with many players - users, ISPs, spammers, technologists, legal

322 views • 17 slides
Test automation Whenever we alter our code, we should re-test Our collections of test cases

Test automation Whenever we alter our code, we should re-test Our collections of test cases

Test automation Whenever we alter our code, we should re-test Our collections of test cases can be substantial, and each test case can be large: manual testing is slow, error-prone Ideally, we would like a tool or script that lets us

186 views • 7 slides
Building Reusable Test Collections Ellen M. Voorhees 1 Test Collections Evaluate search

Building Reusable Test Collections Ellen M. Voorhees 1 Test Collections Evaluate search

Building Reusable Test Collections Ellen M. Voorhees 1 Test Collections Evaluate search effectiveness using test collections set of documents set of questions relevance judgments 20 Relevance judgments Number Relevant

258 views • 23 slides
Spam Fighting at CERN 28 April 2004 Emmanuel Ormancey 1 What is Spam ? What is Spam ? Spam

Spam Fighting at CERN 28 April 2004 Emmanuel Ormancey 1 What is Spam ? What is Spam ? Spam

Spam Fighting at CERN 28 April 2004 Emmanuel Ormancey 1 What is Spam ? What is Spam ? Spam is the friendly name given to unsolicited mail everyone receives in the mailbox. Comes from a Monty Python sketch, where in a caf everything

505 views • 13 slides
Opinion Spam and Analysis NITIN JINDAL & BING LIU, WSDM 08 UIUC Opinion/Review Spam All

Opinion Spam and Analysis NITIN JINDAL & BING LIU, WSDM 08 UIUC Opinion/Review Spam All

Opinion Spam and Analysis NITIN JINDAL & BING LIU, WSDM 08 UIUC Opinion/Review Spam All spam is spam but some spam is more spam than others Opinion spam similar to web spam or email spam in intent, but different in form / content Web

912 views • 38 slides
The CAN-SPAM Act of 2003 D E C E M B E R 2 0 0 3 THE CAN-SPAM ACT OF 2003 Status of Legislation

The CAN-SPAM Act of 2003 D E C E M B E R 2 0 0 3 THE CAN-SPAM ACT OF 2003 Status of Legislation

The CAN-SPAM Act of 2003 D E C E M B E R 2 0 0 3 THE CAN-SPAM ACT OF 2003 Status of Legislation When Congress returns from recess on December 8, 2003, it will finish its work on the CAN-SPAM Act of 2003 (Controlling the Assault of

323 views • 4 slides
Link Spam Alliances Zoltn Gyngyi Hector Garcia-Molina Class List Spam 101 Intro to

Link Spam Alliances Zoltn Gyngyi Hector Garcia-Molina Class List Spam 101 Intro to

Link Spam Alliances Zoltn Gyngyi Hector Garcia-Molina Class List Spam 101 Intro to web spam Spam 221 Spamming PageRank Spam farm model Optimal farm structure Alliances of two farms Larger alliances Spam

876 views • 30 slides
Web Spam Dr. Marc Spaniol Dr. Marc Spaniol Saarbrcken, June 24, 2010 Databases and

Web Spam Dr. Marc Spaniol Dr. Marc Spaniol Saarbrcken, June 24, 2010 Databases and

Web Dynamics Web Spam Web Spam Dr. Marc Spaniol Dr. Marc Spaniol Saarbrcken, June 24, 2010 Databases and Information Systems Prof. Dr. G. Weikum MPII-Sp-0710-1/49 Agenda Web spam - Why and what? Web Spam - Spam taxonomy

669 views • 49 slides
Preventing Use-after-free with Dangling Pointers Nullification Byoungyoung Lee , Chengyu Song,

Preventing Use-after-free with Dangling Pointers Nullification Byoungyoung Lee , Chengyu Song,

Preventing Use-after-free with Dangling Pointers Nullification Byoungyoung Lee , Chengyu Song, Yeongjin Jang Tielei Wang, Taesoo Kim, Long Lu, Wenke Lee Georgia Institute of Technology Stony Brook University Emerging Threat: Use-after-free

923 views • 43 slides
Web Spam Marc Spaniol Marc Spaniol Saarbrcken, July 23, 2009 Databases and Information

Web Spam Marc Spaniol Marc Spaniol Saarbrcken, July 23, 2009 Databases and Information

Web Dynamics Web Spam Web Spam Marc Spaniol Marc Spaniol Saarbrcken, July 23, 2009 Databases and Information Systems Prof. Dr. G. Weikum MPII-Sp-0709-1/49 Agenda Web spam - Why and what? Web Spam - Spam taxonomy Overview

806 views • 49 slides
Naive Bayes case study Training set: 10,000 emails that are either SPAM or HAM Testing set:

Naive Bayes case study Training set: 10,000 emails that are either SPAM or HAM Testing set:

Naive Bayes case study Training set: 10,000 emails that are either SPAM or HAM Testing set: 1,000 additional emails Train a Naive Bayes classifier on (a subset of) the training set Predict SPAM/HAM on the test set and compute

775 views • 13 slides
Detecting Product Review Spammers using Rating Behaviors Itay Dressler What is Spam? Why

Detecting Product Review Spammers using Rating Behaviors Itay Dressler What is Spam? Why

Detecting Product Review Spammers using Rating Behaviors Itay Dressler What is Spam? Why should you care? How to detect Spam? 2 What is Spam? What is Spam? All forms of malicious manipulation of user generated data so as to

943 views • 56 slides
Web Spam Know Your Neighbors: Web Spam Detection using the Web Topology Presenter: Sadia Masood

Web Spam Know Your Neighbors: Web Spam Detection using the Web Topology Presenter: Sadia Masood

Seminar: Future Of Web Search University of Saarland Web Spam Know Your Neighbors: Web Spam Detection using the Web Topology Presenter: Sadia Masood Tutor : Klaus Berberich Date : 17-Jan-2008 The Agenda Focus of the First Paper

1.22k views • 53 slides
spam, ham and other food or how to distribute spam to 100k email addresses Who am I? Debian

spam, ham and other food or how to distribute spam to 100k email addresses Who am I? Debian

spam, ham and other food or how to distribute spam to 100k email addresses Who am I? Debian Developer since 2003 Listmaster Alioth Admin Maintainer of packages like iproute, icinga, nagios and a lot of other useless stu ff

219 views • 20 slides
Collections Objectives Explore collections in System.Collections namespace memory

Collections Objectives Explore collections in System.Collections namespace memory

Collections Objectives Explore collections in System.Collections namespace memory management containment testing sorting traversal 2 Collection A collection stores a group of objects Provides operations to

499 views • 32 slides
Machine Learning in Spam Filtering A Crash Course in ML Konstantin Tretyakov kt@ut.ee

Machine Learning in Spam Filtering A Crash Course in ML Konstantin Tretyakov kt@ut.ee

Machine Learning in Spam Filtering A Crash Course in ML Konstantin Tretyakov kt@ut.ee Institute of Computer Science, University of Tartu Overview Spam is Evil ML for Spam Filtering: General Idea, Problems. Some Algorithms Nave Bayesian

391 views • 35 slides
Baseline A Library for Rapid Modeling, Experimentation and Development of Deep Learning

Baseline A Library for Rapid Modeling, Experimentation and Development of Deep Learning

Baseline A Library for Rapid Modeling, Experimentation and Development of Deep Learning Algorithms targeting NLP Daniel Pressel, Sagnik Ray Choudhury, Brian Lester, Yanjie Zhao, Matt Barta NLP OSS Workshop @ ACL 2018 Baseline: A Deep NLP

441 views • 20 slides
CSE543 - Introduction to Computer and Network Security Module: System Vulnerabilities Professor

CSE543 - Introduction to Computer and Network Security Module: System Vulnerabilities Professor

1.66k views • 27 slides
Statistical Natural Language Parsing Gerald Penn [based on slides by Christopher Manning]

Statistical Natural Language Parsing Gerald Penn [based on slides by Christopher Manning]

Statistical Natural Language Parsing Gerald Penn [based on slides by Christopher Manning] Parsing examples Fed raises interest rates Parsing in the early 1990s The parsers produced detailed, linguistically rich representations

849 views • 46 slides
TRENDSACTIVE x BrandTrust Where Brands meet Human Context to thrive Do you ... ... have

TRENDSACTIVE x BrandTrust Where Brands meet Human Context to thrive Do you ... ... have

TRENDSACTIVE x BrandTrust Where Brands meet Human Context to thrive Do you ... ... have lost the feeling of who your real customer is and what ... is important to them in life? ... have an uninspiring mission or vision that doesnt

514 views • 9 slides
A Unified Contextual Bandit Framework for Long- and Short-Term Recommendations Maryam Tavakol and

A Unified Contextual Bandit Framework for Long- and Short-Term Recommendations Maryam Tavakol and

A Unified Contextual Bandit Framework for Long- and Short-Term Recommendations Maryam Tavakol and Ulf Brefeld {tavakol,brefeld}@leuphana.de Skopje - Sep 21, 2017 Recommendation Tavakol & Brefeld, Leuphana University Lneburg 2/22

513 views • 29 slides
DEMYSTIFYING DATAJOURNALISM In collaboration with Singapore Press Club singapore The DJA are

DEMYSTIFYING DATAJOURNALISM In collaboration with Singapore Press Club singapore The DJA are

singapore FIRST HACKS HACKERS EVENT IN SINGAPORE DEMYSTIFYING DATAJOURNALISM In collaboration with Singapore Press Club singapore The DJA are global awards that recognise excellence in data journalism, organised by the Global Editors

363 views • 11 slides
SMART CITIES Cities of the futur? By Emmanuel Eveno LISST-CIEU, 18 juillet 2016 PROLEGOMENA

SMART CITIES Cities of the futur? By Emmanuel Eveno LISST-CIEU, 18 juillet 2016 PROLEGOMENA

Workshop on Smart and Sustainable City Toulouse, le 18 juillet 2016 SMART CITIES Cities of the futur? By Emmanuel Eveno LISST-CIEU, 18 juillet 2016 PROLEGOMENA Attempt of definition of the object A NEW THEMATIC : an expression in the

1.07k views • 47 slides
Web Dynamics Part 1 - Introduction 1.1 Dimensions of dynamics in the Web 1.2 Application

Web Dynamics Part 1 - Introduction 1.1 Dimensions of dynamics in the Web 1.2 Application

Web Dynamics Part 1 - Introduction 1.1 Dimensions of dynamics in the Web 1.2 Application examples Summer Term 2010 Web Dynamics 1-1 Why Web Dynamics? From Wikipedia: In physics the term dynamics customarily refers to the time evolution of

440 views • 42 slides

More recommend