Towards adaptive Web sites: conceptual framework and case study
Identifieur interne : 000063 ( Istex/Corpus ); précédent : 000062; suivant : 000064Towards adaptive Web sites: conceptual framework and case study
Auteurs : Mike Perkowitz ; Oren EtzioniSource :
- Computer Networks [ 1389-1286 ] ; 1999.
English descriptors
- Teeft :
- Access logs, Adaptive, Algorithm, Apriori, Apriori data mining algorithm, Association rules, Audio samples, Average visit percentage, Candidate clusters, Candidate link sets, Case study, Clique, Cluster, Cluster contents, Cluster mining, Computer science, Conf, Customization, Data mining, Design space, Elsevier science, Etzioni, Etzioni computer networks, Frequent sets, Future work, Graph algorithms, Graph representation, High quality clusters, Human judges, Human webmaster, Humanauthored index pages, Index page, Index page synthesis, Index page synthesis problem, Index pages, Information processing, Information retrieval, Intelligent user interfaces, Joint conf, Large databases, Large number, Links, Matrix, Maximal cliques, Mining association rules, Music machines, Nding clusters, Oren etzioni, Original design, Other hand, Other projects, Other users, Overlap, Overlap reduction, Page views, Pagegather, Pagegather algorithm, Pagegather link, Particular topic, Particular type, Perkowitz, Pgcc, Pgclique, Popular links, Presentation agent, Previous work, Proc, Quality threshold, Reduction step, Research interests, Server, Server logs, Several subproblems, Similar clusters, Similar tastes, Similarity matrix, Single session, Single visit, Small number, Sparse graph, Test data, Training data, Unlinked pages, User, User access patterns, User visits, Visit percentage, Visitor access patterns, Vldb conference, Webmaster, Webwatcher.
Abstract
The creation of a complex Web site is a thorny problem in user interface design. In this paper we explore the notion of adaptive Web sites: sites that semi-automatically improve their organization and presentation by learning from visitor access patterns. It is easy to imagine and implement Web sites that offer shortcuts to popular pages. Are more sophisticated adaptive Web sites feasible? What degree of automation can we achieve? To address the questions above, we describe the design space of adaptive Web sites and consider a case study: the problem of synthesizing new index pages that facilitate navigation of a Web site. We present the PageGather algorithm, which automatically identifies candidate link sets to include in index pages based on user access logs. We demonstrate experimentally that PageGather outperforms the Apriori data mining algorithm on this task. In addition, we compare PageGather's link sets to pre-existing, human-authored index pages.
Url:
DOI: 10.1016/S1389-1286(99)00017-1
Links to Exploration step
ISTEX:05EC28495743552A2AEDA47BE70E386D1DCB64E3Le document en format XML
<record><TEI wicri:istexFullTextTei="biblStruct"><teiHeader><fileDesc><titleStmt><title>Towards adaptive Web sites: conceptual framework and case study</title><author><name sortKey="Perkowitz, Mike" sort="Perkowitz, Mike" uniqKey="Perkowitz M" first="Mike" last="Perkowitz">Mike Perkowitz</name><affiliation><mods:affiliation>Department of Computer Science and Engineering, Box 352350, University of Washington, Seattle, WA 98195, USA</mods:affiliation></affiliation></author><author><name sortKey="Etzioni, Oren" sort="Etzioni, Oren" uniqKey="Etzioni O" first="Oren" last="Etzioni">Oren Etzioni</name><affiliation><mods:affiliation>Department of Computer Science and Engineering, Box 352350, University of Washington, Seattle, WA 98195, USA</mods:affiliation></affiliation></author></titleStmt><publicationStmt><idno type="wicri:source">ISTEX</idno><idno type="RBID">ISTEX:05EC28495743552A2AEDA47BE70E386D1DCB64E3</idno><date when="1999" year="1999">1999</date><idno type="doi">10.1016/S1389-1286(99)00017-1</idno><idno type="url">https://api.istex.fr/document/05EC28495743552A2AEDA47BE70E386D1DCB64E3/fulltext/pdf</idno><idno type="wicri:Area/Istex/Corpus">000063</idno><idno type="wicri:explorRef" wicri:stream="Istex" wicri:step="Corpus" wicri:corpus="ISTEX">000063</idno></publicationStmt><sourceDesc><biblStruct><analytic><title level="a">Towards adaptive Web sites: conceptual framework and case study</title><author><name sortKey="Perkowitz, Mike" sort="Perkowitz, Mike" uniqKey="Perkowitz M" first="Mike" last="Perkowitz">Mike Perkowitz</name><affiliation><mods:affiliation>Department of Computer Science and Engineering, Box 352350, University of Washington, Seattle, WA 98195, USA</mods:affiliation></affiliation></author><author><name sortKey="Etzioni, Oren" sort="Etzioni, Oren" uniqKey="Etzioni O" first="Oren" last="Etzioni">Oren Etzioni</name><affiliation><mods:affiliation>Department of Computer Science and Engineering, Box 352350, University of Washington, Seattle, WA 98195, USA</mods:affiliation></affiliation></author></analytic><monogr></monogr><series><title level="j">Computer Networks</title><title level="j" type="abbrev">COMPNW</title><idno type="ISSN">1389-1286</idno><imprint><publisher>ELSEVIER</publisher><date type="published" when="1999">1999</date><biblScope unit="volume">31</biblScope><biblScope unit="issue">11–16</biblScope><biblScope unit="page" from="1245">1245</biblScope><biblScope unit="page" to="1258">1258</biblScope></imprint><idno type="ISSN">1389-1286</idno></series></biblStruct></sourceDesc><seriesStmt><idno type="ISSN">1389-1286</idno></seriesStmt></fileDesc><profileDesc><textClass><keywords scheme="Teeft" xml:lang="en"><term>Access logs</term><term>Adaptive</term><term>Algorithm</term><term>Apriori</term><term>Apriori data mining algorithm</term><term>Association rules</term><term>Audio samples</term><term>Average visit percentage</term><term>Candidate clusters</term><term>Candidate link sets</term><term>Case study</term><term>Clique</term><term>Cluster</term><term>Cluster contents</term><term>Cluster mining</term><term>Computer science</term><term>Conf</term><term>Customization</term><term>Data mining</term><term>Design space</term><term>Elsevier science</term><term>Etzioni</term><term>Etzioni computer networks</term><term>Frequent sets</term><term>Future work</term><term>Graph algorithms</term><term>Graph representation</term><term>High quality clusters</term><term>Human judges</term><term>Human webmaster</term><term>Humanauthored index pages</term><term>Index page</term><term>Index page synthesis</term><term>Index page synthesis problem</term><term>Index pages</term><term>Information processing</term><term>Information retrieval</term><term>Intelligent user interfaces</term><term>Joint conf</term><term>Large databases</term><term>Large number</term><term>Links</term><term>Matrix</term><term>Maximal cliques</term><term>Mining association rules</term><term>Music machines</term><term>Nding clusters</term><term>Oren etzioni</term><term>Original design</term><term>Other hand</term><term>Other projects</term><term>Other users</term><term>Overlap</term><term>Overlap reduction</term><term>Page views</term><term>Pagegather</term><term>Pagegather algorithm</term><term>Pagegather link</term><term>Particular topic</term><term>Particular type</term><term>Perkowitz</term><term>Pgcc</term><term>Pgclique</term><term>Popular links</term><term>Presentation agent</term><term>Previous work</term><term>Proc</term><term>Quality threshold</term><term>Reduction step</term><term>Research interests</term><term>Server</term><term>Server logs</term><term>Several subproblems</term><term>Similar clusters</term><term>Similar tastes</term><term>Similarity matrix</term><term>Single session</term><term>Single visit</term><term>Small number</term><term>Sparse graph</term><term>Test data</term><term>Training data</term><term>Unlinked pages</term><term>User</term><term>User access patterns</term><term>User visits</term><term>Visit percentage</term><term>Visitor access patterns</term><term>Vldb conference</term><term>Webmaster</term><term>Webwatcher</term></keywords></textClass><langUsage><language ident="en">en</language></langUsage></profileDesc></teiHeader><front><div type="abstract" xml:lang="en">The creation of a complex Web site is a thorny problem in user interface design. In this paper we explore the notion of adaptive Web sites: sites that semi-automatically improve their organization and presentation by learning from visitor access patterns. It is easy to imagine and implement Web sites that offer shortcuts to popular pages. Are more sophisticated adaptive Web sites feasible? What degree of automation can we achieve? To address the questions above, we describe the design space of adaptive Web sites and consider a case study: the problem of synthesizing new index pages that facilitate navigation of a Web site. We present the PageGather algorithm, which automatically identifies candidate link sets to include in index pages based on user access logs. We demonstrate experimentally that PageGather outperforms the Apriori data mining algorithm on this task. In addition, we compare PageGather's link sets to pre-existing, human-authored index pages.</div></front></TEI><istex><corpusName>elsevier</corpusName><keywords><teeft><json:string>pagegather</json:string><json:string>pgclique</json:string><json:string>adaptive</json:string><json:string>webmaster</json:string><json:string>algorithm</json:string><json:string>index pages</json:string><json:string>perkowitz</json:string><json:string>etzioni</json:string><json:string>apriori</json:string><json:string>etzioni computer networks</json:string><json:string>overlap reduction</json:string><json:string>customization</json:string><json:string>pgcc</json:string><json:string>server</json:string><json:string>proc</json:string><json:string>candidate link sets</json:string><json:string>pagegather algorithm</json:string><json:string>cluster mining</json:string><json:string>webwatcher</json:string><json:string>user</json:string><json:string>human webmaster</json:string><json:string>conf</json:string><json:string>music machines</json:string><json:string>page views</json:string><json:string>matrix</json:string><json:string>index page</json:string><json:string>index page synthesis</json:string><json:string>similarity matrix</json:string><json:string>data mining</json:string><json:string>average visit percentage</json:string><json:string>visitor access patterns</json:string><json:string>future work</json:string><json:string>maximal cliques</json:string><json:string>case study</json:string><json:string>test data</json:string><json:string>computer science</json:string><json:string>cluster</json:string><json:string>index page synthesis problem</json:string><json:string>design space</json:string><json:string>user visits</json:string><json:string>unlinked pages</json:string><json:string>mining association rules</json:string><json:string>user access patterns</json:string><json:string>particular topic</json:string><json:string>access logs</json:string><json:string>information retrieval</json:string><json:string>candidate clusters</json:string><json:string>previous work</json:string><json:string>vldb conference</json:string><json:string>small number</json:string><json:string>cluster contents</json:string><json:string>other users</json:string><json:string>clique</json:string><json:string>links</json:string><json:string>overlap</json:string><json:string>training data</json:string><json:string>popular links</json:string><json:string>nding clusters</json:string><json:string>graph algorithms</json:string><json:string>sparse graph</json:string><json:string>single visit</json:string><json:string>presentation agent</json:string><json:string>similar clusters</json:string><json:string>high quality clusters</json:string><json:string>other projects</json:string><json:string>similar tastes</json:string><json:string>other hand</json:string><json:string>quality threshold</json:string><json:string>large number</json:string><json:string>frequent sets</json:string><json:string>audio samples</json:string><json:string>several subproblems</json:string><json:string>server logs</json:string><json:string>graph representation</json:string><json:string>reduction step</json:string><json:string>particular type</json:string><json:string>visit percentage</json:string><json:string>humanauthored index pages</json:string><json:string>single session</json:string><json:string>human judges</json:string><json:string>pagegather link</json:string><json:string>original design</json:string><json:string>large databases</json:string><json:string>elsevier science</json:string><json:string>association rules</json:string><json:string>apriori data mining algorithm</json:string><json:string>oren etzioni</json:string><json:string>joint conf</json:string><json:string>intelligent user interfaces</json:string><json:string>information processing</json:string><json:string>research interests</json:string></teeft></keywords><author><json:item><name>Mike Perkowitz</name><affiliations><json:string>Department of Computer Science and Engineering, Box 352350, University of Washington, Seattle, WA 98195, USA</json:string></affiliations></json:item><json:item><name>Oren Etzioni</name><affiliations><json:string>Department of Computer Science and Engineering, Box 352350, University of Washington, Seattle, WA 98195, USA</json:string></affiliations></json:item></author><subject><json:item><lang><json:string>eng</json:string></lang><value>Adaptive</value></json:item><json:item><lang><json:string>eng</json:string></lang><value>Clustering</value></json:item><json:item><lang><json:string>eng</json:string></lang><value>Data mining</value></json:item></subject><language><json:string>eng</json:string></language><originalGenre><json:string>Full-length article</json:string></originalGenre><abstract>The creation of a complex Web site is a thorny problem in user interface design. In this paper we explore the notion of adaptive Web sites: sites that semi-automatically improve their organization and presentation by learning from visitor access patterns. It is easy to imagine and implement Web sites that offer shortcuts to popular pages. Are more sophisticated adaptive Web sites feasible? What degree of automation can we achieve? To address the questions above, we describe the design space of adaptive Web sites and consider a case study: the problem of synthesizing new index pages that facilitate navigation of a Web site. We present the PageGather algorithm, which automatically identifies candidate link sets to include in index pages based on user access logs. We demonstrate experimentally that PageGather outperforms the Apriori data mining algorithm on this task. In addition, we compare PageGather's link sets to pre-existing, human-authored index pages.</abstract><qualityIndicators><score>6.788</score><pdfVersion>1.2</pdfVersion><pdfPageSize>546 x 746 pts</pdfPageSize><refBibsNative>true</refBibsNative><keywordCount>3</keywordCount><abstractCharCount>969</abstractCharCount><pdfWordCount>7725</pdfWordCount><pdfCharCount>45527</pdfCharCount><pdfPageCount>14</pdfPageCount><abstractWordCount>149</abstractWordCount></qualityIndicators><title>Towards adaptive Web sites: conceptual framework and case study</title><pii><json:string>S1389-1286(99)00017-1</json:string></pii><genre><json:string>research-article</json:string></genre><host><title>Computer Networks</title><language><json:string>unknown</json:string></language><publicationDate>1999</publicationDate><issn><json:string>1389-1286</json:string></issn><pii><json:string>S1389-1286(00)X0008-4</json:string></pii><volume>31</volume><issue>11–16</issue><pages><first>1245</first><last>1258</last></pages><genre><json:string>journal</json:string></genre></host><categories><wos><json:string>science</json:string><json:string>telecommunications</json:string><json:string>engineering, electrical & electronic</json:string><json:string>computer science, information systems</json:string><json:string>computer science, hardware & architecture</json:string></wos><scienceMetrix><json:string>applied sciences</json:string><json:string>information & communication technologies</json:string><json:string>networking & telecommunications</json:string></scienceMetrix><inist><json:string>sciences humaines et sociales</json:string></inist></categories><publicationDate>1999</publicationDate><copyrightDate>1999</copyrightDate><doi><json:string>10.1016/S1389-1286(99)00017-1</json:string></doi><id>05EC28495743552A2AEDA47BE70E386D1DCB64E3</id><score>1</score><fulltext><json:item><extension>pdf</extension><original>true</original><mimetype>application/pdf</mimetype><uri>https://api.istex.fr/document/05EC28495743552A2AEDA47BE70E386D1DCB64E3/fulltext/pdf</uri></json:item><json:item><extension>zip</extension><original>false</original><mimetype>application/zip</mimetype><uri>https://api.istex.fr/document/05EC28495743552A2AEDA47BE70E386D1DCB64E3/fulltext/zip</uri></json:item><istex:fulltextTEI uri="https://api.istex.fr/document/05EC28495743552A2AEDA47BE70E386D1DCB64E3/fulltext/tei"><teiHeader><fileDesc><titleStmt><title level="a">Towards adaptive Web sites: conceptual framework and case study</title></titleStmt><publicationStmt><authority>ISTEX</authority><publisher>ELSEVIER</publisher><availability><p>ELSEVIER</p></availability><date>1999</date></publicationStmt><notesStmt><note type="content">Fig. 1: Typical user access logs, these from a computer science Web site. Each entry corresponds to a single request to the server and includes originating machine, time, and URL requested. Note the series of accesses from each of two users (one from SFSU, one from UMN).</note><note type="content">Fig. 2: (a) A candidate cluster to be presented to the Webmaster for approval and naming. (b) How the final page would appear at the site, properly named and formatted.</note><note type="content">Fig. 3: The performance of PGCLIQUE and PGCC using raw ranked-cluster output with no overlap reduction. Although PGCLIQUE apparently performs much better, its clusters are all variations on the same basic set of pages.</note><note type="content">Fig. 4: The performance of PGCLIQUE and PGCC using overlap reduction. PGCLIQUE performs much better on the top three clusters and generally better overall.</note><note type="content">Fig. 5: Comparing the performance of PGCLIQUE with overlap reduction to PGCLIQUE with cluster merging. Reduction performs better on the first three clusters, but the two variants are otherwise comparable.</note><note type="content">Fig. 6: The performance of PGCLIQUE with overlap reduction compared with the Apriori algorithm, both with and without overlap reduction. PGCLIQUE performs significantly better than both variations. Note that Apriori finds only two distinct clusters when overlap reduction is applied.</note><note type="content">Fig. 7: The performance of PGCLIQUE using overlap reduction as compared to the performance of clusters based on human-authored index pages at the Music Machines Web site. Clusters found by PGCLIQUE perform significantly better than the existing index pages.</note><note type="content">Table 1: Clusters found by PGCLIQUE with overlap reduction</note></notesStmt><sourceDesc><biblStruct type="inbook"><analytic><title level="a">Towards adaptive Web sites: conceptual framework and case study</title><author xml:id="author-0000"><persName><forename type="first">Mike</forename><surname>Perkowitz</surname></persName><note type="biography">E-mail: {map,etzioni}@cs.washington.edu</note><note type="correspondence"><p>Corresponding author.</p></note><affiliation>Department of Computer Science and Engineering, Box 352350, University of Washington, Seattle, WA 98195, USA</affiliation></author><author xml:id="author-0001"><persName><forename type="first">Oren</forename><surname>Etzioni</surname></persName><note type="biography">E-mail: {map,etzioni}@cs.washington.edu</note><affiliation>Department of Computer Science and Engineering, Box 352350, University of Washington, Seattle, WA 98195, USA</affiliation></author><idno type="istex">05EC28495743552A2AEDA47BE70E386D1DCB64E3</idno><idno type="DOI">10.1016/S1389-1286(99)00017-1</idno><idno type="PII">S1389-1286(99)00017-1</idno></analytic><monogr><title level="j">Computer Networks</title><title level="j" type="abbrev">COMPNW</title><idno type="pISSN">1389-1286</idno><idno type="PII">S1389-1286(00)X0008-4</idno><imprint><publisher>ELSEVIER</publisher><date type="published" when="1999"></date><biblScope unit="volume">31</biblScope><biblScope unit="issue">11–16</biblScope><biblScope unit="page" from="1245">1245</biblScope><biblScope unit="page" to="1258">1258</biblScope></imprint></monogr></biblStruct></sourceDesc></fileDesc><profileDesc><creation><date>1999</date></creation><langUsage><language ident="en">en</language></langUsage><abstract xml:lang="en"><p>The creation of a complex Web site is a thorny problem in user interface design. In this paper we explore the notion of adaptive Web sites: sites that semi-automatically improve their organization and presentation by learning from visitor access patterns. It is easy to imagine and implement Web sites that offer shortcuts to popular pages. Are more sophisticated adaptive Web sites feasible? What degree of automation can we achieve? To address the questions above, we describe the design space of adaptive Web sites and consider a case study: the problem of synthesizing new index pages that facilitate navigation of a Web site. We present the PageGather algorithm, which automatically identifies candidate link sets to include in index pages based on user access logs. We demonstrate experimentally that PageGather outperforms the Apriori data mining algorithm on this task. In addition, we compare PageGather's link sets to pre-existing, human-authored index pages.</p></abstract><textClass><keywords scheme="keyword"><list><head>Keywords</head><item><term>Adaptive</term></item><item><term>Clustering</term></item><item><term>Data mining</term></item></list></keywords></textClass></profileDesc><revisionDesc><change when="1999">Published</change></revisionDesc></teiHeader></istex:fulltextTEI><json:item><extension>txt</extension><original>false</original><mimetype>text/plain</mimetype><uri>https://api.istex.fr/document/05EC28495743552A2AEDA47BE70E386D1DCB64E3/fulltext/txt</uri></json:item></fulltext><metadata><istex:metadataXml wicri:clean="Elsevier, elements deleted: ce:floats; body; tail"><istex:xmlDeclaration>version="1.0" encoding="utf-8"</istex:xmlDeclaration><istex:docType PUBLIC="-//ES//DTD journal article DTD version 4.5.2//EN//XML" URI="art452.dtd" name="istex:docType"><istex:entity SYSTEM="gr1" NDATA="IMAGE" name="gr1"></istex:entity><istex:entity SYSTEM="gr2" NDATA="IMAGE" name="gr2"></istex:entity><istex:entity SYSTEM="gr3" NDATA="IMAGE" name="gr3"></istex:entity><istex:entity SYSTEM="gr4" NDATA="IMAGE" name="gr4"></istex:entity><istex:entity SYSTEM="gr5" NDATA="IMAGE" name="gr5"></istex:entity><istex:entity SYSTEM="gr6" NDATA="IMAGE" name="gr6"></istex:entity><istex:entity SYSTEM="gr7" NDATA="IMAGE" name="gr7"></istex:entity><istex:entity SYSTEM="fx2" NDATA="IMAGE" name="fx2"></istex:entity><istex:entity SYSTEM="fx1" NDATA="IMAGE" name="fx1"></istex:entity></istex:docType><istex:document><converted-article version="4.5.2" docsubtype="fla"><item-info><jid>COMPNW</jid><aid>2143</aid><ce:pii>S1389-1286(99)00017-1</ce:pii><ce:doi>10.1016/S1389-1286(99)00017-1</ce:doi><ce:copyright year="1999" type="unknown"></ce:copyright></item-info><head><ce:title>Towards adaptive Web sites: conceptual framework and case study</ce:title><ce:author-group><ce:author biographyid="BIO1"><ce:given-name>Mike</ce:given-name><ce:surname>Perkowitz</ce:surname><ce:cross-ref refid="FN1"><ce:sup>1</ce:sup></ce:cross-ref><ce:cross-ref refid="CORR1">*</ce:cross-ref><ce:cross-ref refid="FN1"><ce:sup>1</ce:sup></ce:cross-ref></ce:author><ce:author><ce:given-name>Oren</ce:given-name><ce:surname>Etzioni</ce:surname><ce:cross-ref refid="FN1"><ce:sup>1</ce:sup></ce:cross-ref></ce:author><ce:affiliation><ce:textfn>Department of Computer Science and Engineering, Box 352350, University of Washington, Seattle, WA 98195, USA</ce:textfn></ce:affiliation><ce:correspondence id="CORR1"><ce:label>*</ce:label><ce:text>Corresponding author.</ce:text></ce:correspondence><ce:footnote id="FN1"><ce:label>1</ce:label><ce:note-para>E-mail: {map,etzioni}@cs.washington.edu</ce:note-para></ce:footnote></ce:author-group><ce:abstract><ce:section-title>Abstract</ce:section-title><ce:abstract-sec><ce:simple-para>The creation of a complex Web site is a thorny problem in user interface design. In this paper we explore the notion of <ce:bold>adaptive Web sites</ce:bold>: sites that semi-automatically improve their organization and presentation by learning from visitor access patterns. It is easy to imagine and implement Web sites that offer shortcuts to popular pages. Are more sophisticated adaptive Web sites feasible? What degree of automation can we achieve?</ce:simple-para><ce:simple-para>To address the questions above, we describe the design space of adaptive Web sites and consider a case study: the problem of synthesizing new <ce:italic>index pages</ce:italic> that facilitate navigation of a Web site. We present the PageGather algorithm, which automatically identifies candidate link sets to include in index pages based on user access logs. We demonstrate experimentally that PageGather outperforms the Apriori data mining algorithm on this task. In addition, we compare PageGather's link sets to pre-existing, human-authored index pages.</ce:simple-para></ce:abstract-sec></ce:abstract><ce:keywords class="keyword"><ce:section-title>Keywords</ce:section-title><ce:keyword><ce:text>Adaptive</ce:text></ce:keyword><ce:keyword><ce:text>Clustering</ce:text></ce:keyword><ce:keyword><ce:text>Data mining</ce:text></ce:keyword></ce:keywords></head></converted-article></istex:document></istex:metadataXml><mods version="3.6"><titleInfo><title>Towards adaptive Web sites: conceptual framework and case study</title></titleInfo><titleInfo type="alternative" contentType="CDATA"><title>Towards adaptive Web sites: conceptual framework and case study</title></titleInfo><name type="personal"><namePart type="given">Mike</namePart><namePart type="family">Perkowitz</namePart><affiliation>Department of Computer Science and Engineering, Box 352350, University of Washington, Seattle, WA 98195, USA</affiliation><description>Corresponding author.</description><description>E-mail: {map,etzioni}@cs.washington.edu</description><role><roleTerm type="text">author</roleTerm></role></name><name type="personal"><namePart type="given">Oren</namePart><namePart type="family">Etzioni</namePart><affiliation>Department of Computer Science and Engineering, Box 352350, University of Washington, Seattle, WA 98195, USA</affiliation><description>E-mail: {map,etzioni}@cs.washington.edu</description><role><roleTerm type="text">author</roleTerm></role></name><typeOfResource>text</typeOfResource><genre type="research-article" displayLabel="Full-length article"></genre><originInfo><publisher>ELSEVIER</publisher><dateIssued encoding="w3cdtf">1999</dateIssued><copyrightDate encoding="w3cdtf">1999</copyrightDate></originInfo><language><languageTerm type="code" authority="iso639-2b">eng</languageTerm><languageTerm type="code" authority="rfc3066">en</languageTerm></language><physicalDescription><internetMediaType>text/html</internetMediaType></physicalDescription><abstract lang="en">The creation of a complex Web site is a thorny problem in user interface design. In this paper we explore the notion of adaptive Web sites: sites that semi-automatically improve their organization and presentation by learning from visitor access patterns. It is easy to imagine and implement Web sites that offer shortcuts to popular pages. Are more sophisticated adaptive Web sites feasible? What degree of automation can we achieve? To address the questions above, we describe the design space of adaptive Web sites and consider a case study: the problem of synthesizing new index pages that facilitate navigation of a Web site. We present the PageGather algorithm, which automatically identifies candidate link sets to include in index pages based on user access logs. We demonstrate experimentally that PageGather outperforms the Apriori data mining algorithm on this task. In addition, we compare PageGather's link sets to pre-existing, human-authored index pages.</abstract><note type="content">Fig. 1: Typical user access logs, these from a computer science Web site. Each entry corresponds to a single request to the server and includes originating machine, time, and URL requested. Note the series of accesses from each of two users (one from SFSU, one from UMN).</note><note type="content">Fig. 2: (a) A candidate cluster to be presented to the Webmaster for approval and naming. (b) How the final page would appear at the site, properly named and formatted.</note><note type="content">Fig. 3: The performance of PGCLIQUE and PGCC using raw ranked-cluster output with no overlap reduction. Although PGCLIQUE apparently performs much better, its clusters are all variations on the same basic set of pages.</note><note type="content">Fig. 4: The performance of PGCLIQUE and PGCC using overlap reduction. PGCLIQUE performs much better on the top three clusters and generally better overall.</note><note type="content">Fig. 5: Comparing the performance of PGCLIQUE with overlap reduction to PGCLIQUE with cluster merging. Reduction performs better on the first three clusters, but the two variants are otherwise comparable.</note><note type="content">Fig. 6: The performance of PGCLIQUE with overlap reduction compared with the Apriori algorithm, both with and without overlap reduction. PGCLIQUE performs significantly better than both variations. Note that Apriori finds only two distinct clusters when overlap reduction is applied.</note><note type="content">Fig. 7: The performance of PGCLIQUE using overlap reduction as compared to the performance of clusters based on human-authored index pages at the Music Machines Web site. Clusters found by PGCLIQUE perform significantly better than the existing index pages.</note><note type="content">Table 1: Clusters found by PGCLIQUE with overlap reduction</note><subject><genre>Keywords</genre><topic>Adaptive</topic><topic>Clustering</topic><topic>Data mining</topic></subject><relatedItem type="host"><titleInfo><title>Computer Networks</title></titleInfo><titleInfo type="abbreviated"><title>COMPNW</title></titleInfo><genre type="journal">journal</genre><originInfo><dateIssued encoding="w3cdtf">19990517</dateIssued></originInfo><identifier type="ISSN">1389-1286</identifier><identifier type="PII">S1389-1286(00)X0008-4</identifier><part><date>19990517</date><detail type="volume"><number>31</number><caption>vol.</caption></detail><detail type="issue"><number>11–16</number><caption>no.</caption></detail><extent unit="issue pages"><start>1079</start><end>1778</end></extent><extent unit="pages"><start>1245</start><end>1258</end></extent></part></relatedItem><identifier type="istex">05EC28495743552A2AEDA47BE70E386D1DCB64E3</identifier><identifier type="DOI">10.1016/S1389-1286(99)00017-1</identifier><identifier type="PII">S1389-1286(99)00017-1</identifier><recordInfo><recordContentSource>ELSEVIER</recordContentSource></recordInfo></mods></metadata><serie></serie></istex></record>Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Wicri/Sarre/explor/MusicSarreV3/Data/Istex/Corpus
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 000063 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Istex/Corpus/biblio.hfd -nk 000063 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien
|wiki= Wicri/Sarre
|area= MusicSarreV3
|flux= Istex
|étape= Corpus
|type= RBID
|clé= ISTEX:05EC28495743552A2AEDA47BE70E386D1DCB64E3
|texte= Towards adaptive Web sites: conceptual framework and case study
}}
| This area was generated with Dilib version V0.6.33. |  |