搜索引擎技术（PPT讲稿）Web Spam

Web spam ●●●●● 2008.1220·

Web Spam 2008.12.20

●●● ●●●● ●●●●● ●●●● Outline ●●●●● ●●●● o Motivation ● ntroduction to Web Spam ● Web Spam techniques ● Web Spam Detection ● Conclusions

Outline ⚫ Motivation ⚫ Introduction to Web Spam ⚫ Web Spam techniques ⚫ Web Spam Detection ⚫ Conclusions

●●● ●●●● ●●●●● ●●●● Outline ●●●●● ●●●● ● Motivation o Introduction to Web Spam ● Web Spam Techniques ● Web Spam Detection Conclusions

Outline ⚫ Motivation ⚫ Introduction to Web Spam ⚫ Web Spam Techniques ⚫ Web Spam Detection ⚫ Conclusions

●●● ●●●● ●●●●● ●●●● ●●●●● Motivation ●●●● ●●●● Increased exposure on the World Wide Web may yield significant financial gains E-commerce is rapidly growing Projected to $329 billion by 2010: 13% of all US retail sales ● More traffic> more money Large fraction of traffic from Search Engines Increase Search Engine referrals ● Place ads Provide genuinely better content ● Create Web spam

Motivation ⚫ Increased exposure on the World Wide Web may yield significant financial gains ⚫ E-commerce is rapidly growing ⚫ Projected to $329 billion by 2010;13% of all US retail sales ⚫ More traffic → more money ⚫ Large fraction of traffic from Search Engines ⚫ Increase Search Engine referrals: ⚫ Place ads ☺ ⚫ Provide genuinely better content ☺ ⚫ Create Web spam … 

●●● ●●●● ●●●●● ●●●● Outline ●●●●● ●●●● ● Motivation e Introduction to Web Spam ● Web Spam Techniques ● Web Spam Detection Conclusions

Outline ⚫ Motivation ⚫ Introduction to Web Spam ⚫ Web Spam Techniques ⚫ Web Spam Detection ⚫ Conclusions

●●● ●●●● ●●●●● ●●●● Web spam ●●●●● ●●●● eclick for bad credit report repair -interested what is on your credit? Find out today Microsoft Internet Explorer-0 Favontes Toois He'p urls a] Marcs Irks Fle Edt vew Favorites Tods Heb unis () Marcs links xk,o.回回 P Search = 7Fawortes Me色,品k,o回 oPsearch s? Favorites可hdse, Addressahttp://smslugt.net/folowd Addressehttp://www.iosorire.ora/bed-credt-report-repair.heml bad credit report repair 20-voice-yml htm 3-eib-findbyprumarykey-iboss-gl.htm bad credit report repair find out about Dark men simple kinds plants following dark found usually moved bad credit report free credit check I no credit repair, looked girls English were enternet does sites first checking account no Been check cell phone I no credit looking news make earth behind miles were no for bad credit report repair, man watch check loan I check credit 6 htm night find started down what been score federal credit union. it of made it others boys report I credit card no credIt an't were same cause bad credit report repair, and above do beautiful off against saw check i tree cell phone no abc-soap-opera. htm ave alabama bureau credit. Always English order always who open oline come will credit check l online credit check I no credit check abc-soap. htm ess-ms-rdbms htm change online door news every nothing know most should, name later up seen list personal loan I no credit complete known needed equifax credit report. Place felt who do known the feel kind sh an no payday saving. Such winter largest same says such form day draw certain, went sredit check I tree. online ActiveX htm ada-corba htm numbers point part often or, words strong been during people when several man credit check l payday loan no credlt port credit score, Living against use fast everything shone being shows across wind, checkI sheck credit history I acvantage-rdbms. htm our word seen warm living dark who little consumer credit report agency. Food news check your credit rating girl brought house help feet want gave food, system word always no kind following ADHam seemed each trans union credit score identity theft check I canada credit check I lord c all-my-child- soap-opera htm Alair. htm wwcsonline oral usiness cr w irs goy/newsroom/article/, id= 109810.00 htm aliforniacu ora check credi Armiga-OS. htm Mw irs aovfindivduals/article/0 164600htm personal credit check I free nsterdam-corba nbla htm Apache-Tomcat htm bad credit report repair redit check monitor Apple-Webobjects. htm servce i cell check credit no apphcation-custom-development-stte-software.web htm phone plan i canadian credit aspaml htm aspnet-2am htm student I no credit check auto sembly-language hts ATG- Dynamo. htm wareless phone no credit AtL htm

Web Spam

●●● ●●●● ●●●●● ●●●● ●●●●● Defining Web Spam ●●●● Spamming misleading search engines to obtain higher-than-deserved ranking Tanking Relevance Importance

Defining Web Spam Ranking Relevance Importance Spamming = misleading search engines to obtain higher-than-deserved ranking

●●● ●●●● ●●●●● ●●●● ●●●●● Why Web spam is bad ●●●● ● Bad for users o Makes it harder to satisfy information need o Leads to frustrating search experience o Bad for search engines Wastes bandwidth, CPU cycles, storage space o Pollutes corpus(infinite number of spam pages Distorts ranking of results

Why Web Spam is Bad ⚫ Bad for users ⚫ Makes it harder to satisfy information need ⚫ Leads to frustrating search experience ⚫ Bad for search engines ⚫ Wastes bandwidth, CPU cycles, storage space ⚫ Pollutes corpus (infinite number of spam pages!) ⚫ Distorts ranking of results

●●● ●●●● ●●●●● ●●●● Outline ●●●●● ●●●● ● Motivation o Introduction to Web Spam ● Web Spam Techniques ● Web Spam Detection ● Conclusions References

Outline ⚫ Motivation ⚫ Introduction to Web Spam ⚫ Web Spam Techniques ⚫ Web Spam Detection ⚫ Conclusions ⚫ References

●●● ●●●● ●●●●● ●●●● ●●●●● Web spam Techniques ●●●● o Two categories of techniques associated With web spam Boosting Term-based Link-based Hiding

Web Spam Techniques ⚫ Two categories of techniques associated with web spam ⚫ Boosting ▪ Term-based ▪ Link-based ⚫ Hiding