DEVELOPING A PERSONALIZED ARTICLE RETRIEVAL SYSTEM FOR PUBMED

dc.contributor.advisorLi, Cen
dc.contributor.authorPitigala, Sachintha Prasad
dc.contributor.committeememberSeo, Suk
dc.contributor.committeememberWallin, John
dc.contributor.committeememberWu, Qiang
dc.contributor.departmentBasic & Applied Sciencesen_US
dc.date.accessioned2016-08-15T15:03:30Z
dc.date.available2016-08-15T15:03:30Z
dc.date.issued2016-06-21
dc.description.abstractPubMed keyword based search often results in many citations not directly relevant to the
dc.description.abstractuser information need. Personalized Information Retrieval (PIR) systems aim to improve the
dc.description.abstractquality of the retrieval results by letting the users supply information other than keywords.
dc.description.abstractTwo main problems have been identified for the current PIR systems developed for PubMed:
dc.description.abstract(1) requiring the user to supply a large number of citations directly relevant to a search topic,
dc.description.abstractand (2) producing too many search results, with a high percentage being false positives. This
dc.description.abstractstudy developed a Personalized Article Retrieval System (PARS) for PubMed to address
dc.description.abstractthese problems. PARS uses two main approaches to find the relevant citations to the given
dc.description.abstractinformation need: (1) Extending the PubMed Related Article (PMRA) feature and (2) Text
dc.description.abstractclassification based Multi Stage Filtering (MSF) method. Both approaches require only a
dc.description.abstractsmall set of citations from the user, and reduce the search output size by eliminating the
dc.description.abstractfalse-positive citations in the search output. PARS has been experimentally evaluated using
dc.description.abstractthe TREC 2005 dataset, and empirically evaluated by subject experts from the biomedicine
dc.description.abstractfield. Results show the PARS system is able to produce retrieval results of better quality
dc.description.abstractthan the existing PIR systems for PubMed.
dc.description.degreePh.D.
dc.identifier.urihttp://jewlscholar.mtsu.edu/handle/mtsu/4984
dc.publisherMiddle Tennessee State University
dc.subjectMedical Information Retrieval
dc.subjectPubMed
dc.subjectSimilarity Measure
dc.subjectText Classification
dc.subject.umiComputer science
dc.subject.umiStatistics
dc.thesis.degreegrantorMiddle Tennessee State University
dc.thesis.degreelevelDoctoral
dc.titleDEVELOPING A PERSONALIZED ARTICLE RETRIEVAL SYSTEM FOR PUBMED
dc.typeDissertation

Files

Original bundle

Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
Pitigala_mtsu_0170E_10636.pdf
Size:
7.57 MB
Format:
Adobe Portable Document Format