B. D. Davison. Web tra#c logs: An imperfect resource for evaluation. In Ninth Annual Conference of the Internet Society (INET'99), pages 22--25, June 1999.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....from selecting a hyperlink in a document) In Chapter 6 we will additionally show that a large portion of the clicks will be to links on pages that have been requested in the recent past. Thus, for evaluation of many Web algorithms, typical Web usage logs are insu#cient. As we describe elsewhere [Dav99c] standard logs generated by Web servers or proxies do not provide enough data for accurate analysis of user activity. For example, they do not include many of the headers which specify cacheability, or whether the request is to validate a cached copy. Further, Web resources change rapidly [GS96, ....

....as we described in Chapter 3, or telecommunication switch alarms [Wei01] In contrast, the Web provides easy access to the content being provided to the user. Moreover, this content prescribes boundaries within which the user is likely to act. As described in Chapter 5, studies [CP95, TG97, Dav99c] have shown that the user is likely to select a link on the current page between 80 and 90 of the time. Therefore, the set of links shown to the user is a significant guide to what the user is likely to request next. In addition, the content provides possible next steps that historically based ....

[Article contains additional citation context not shown here]

Brian D. Davison. Web tra#c logs: An imperfect resource for evaluation. In Proceedings of the Ninth Annual Conference of the Internet Society (INET'99), June 1999. 278

.... showing that users predominantly (approximately 80 of the time) choose a link from the page being shown to them (as opposed to the using a bookmark or typing a URL by hand) We have seen similar results from analysis of proxy and server logs (measuring the presence of referrers for HTML pages [19]) Therefore, the set of links shown to the user is a significant guide to what the user is likely to request next. In addition, the content provides possible next steps that historically based prediction approaches cannot. A user model based on past experience cannot predict an action previously ....

B. D. Davison. Web tra#c logs: An imperfect resource for evaluation. In Proceedings of the Ninth Annual Conference of the Internet Society (INET'99), June 1999.

.... [BH96] BDH 94] Bra96] wBC95] BCFP98] BCF 99] BGM98] BD96] CDF 98] CFKL95] CI97] CZB98] CGL 98] CC96] Cat92] CP95] CDN 96] CY97] CMPS95] CLM99] CLLM00] Cor96] Cor98] CB95] CB96] CB97] CB98] CBC95] CJ97] Cun97] DHS93] Dan98] Dav99b] Dav99a] Dav99c] Den96] DCW96] Din96] DP96] DMP95] DFKM97a] DFKM97b] DBCK98] Duc99] DMF97] Dyk98] DJD99] CKR99] CKO99] CKZ99] CK99] CK00] EWCS96] FCAB98] FJCL99] FRC98] FRCar] Fel98] FCD 99] FVYI00] FB96] GRC97] GCR97] GCR98] GC97] Gla94] GPB98] Gri97] GB97] GPV98] GNPV98] ....

Brian D. Davison. Web tra#c logs: An imperfect resource for evaluation. In Proceedings of the Ninth Annual Conference of the Internet Society (INET'99), June 1999.

No context found.

B. D. Davison. Web tra#c logs: An imperfect resource for evaluation. In Ninth Annual Conference of the Internet Society (INET'99), pages 22--25, June 1999.

No context found.

B. D. Davison. Web tra#c logs: An imperfect resource for evaluation. In Ninth Annual Conference of the Internet Society (INET'99), pages 22--25, June 1999.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC