I am deeply indebted to Dr. Michael Berry, my major advisor, for his kind guidance and support. I also thank Dr. Susan Dumais, director of the Information Sciences Research Group at Bellcore, for her technical advice. In addition, she graciously allowed us
and,byexaminingtheclustersofinformation,derivingmeaningfulconclusionsfromthedata.Clearly,conceptualinformationretrievalmodels,becauseofthewaytheyrepresentinformationindependentlyoftheactualdataitself(typically,theyrepresentthedataasapointinspaceratherthanasasetofwords),aresuperiortolexicalmatchingtechniqueswhenoneisattemptingtoderivemeaningfromalarge,heterogeneousdataset.
1.2.2DigitalLibraries
Whetherthewords“digitallibrary”evokeimagesofnewmethodsofaccessingandviewinginformationormerelyanelectronicformofthetraditionallibrary,informationretrievaltechnologieswillplayasigni cantroleinthewaytheinformationisorganizedandsearched.Already,severalWorld-WideWebsiteshavebegunindexingcomputersciencetechnicalreports,allowinguserstosearchabstractsofthetechnicalreportsorthefulltextofthetechnicalreportsthemselves[Fox95].Futuristshopedigitallibrarieswillbeabletoprovidegreateraccesstoinformation,increasetheroyaltiespaidtoauthors,andallowinformationtobedynamic(and,therefore,alsocontinuallyup-to-date)ratherthanstatic[Wie95].
1.2.3InformationFiltering
Asinformationbecomesmoreabundantandeasiertodisseminate,peoplewillbecomelessabletohandlethecontinuous owofinformation.Tohelppeoplebetterdealwiththebarrageofinformation,automaticinformation lteringtechniquescanbeusedtodiscardinformationirrelevanttoaperson’sinterestswhileallowingtherestoftheinformationtopassthroughthe http://www.77cn.com.cnrmation lteringisverysimilartoinformationretrieval.Insteadofretrievinginformationrelatedtoauser’squery,though,theusertypicallycreatesapro leofwhatheorshewishestoview,andtheincomingdocumentsarecomparedtothepro le.Documentsthatmatchthepro leareallowedtopassthroughthe lter,whiletheotherdocumentsarediscarded.The
4