There are several problems regarding information retrieval on biomedical information. The common methods for information retrieval tend to fall short when searching in this domain. With the ever increasing amount of information available, researchers have widely agreed on that means to precisely retrieve needed information is vital to use all available knowledge. We have in an effort to increase the precision of retrieval within biomedical information created an approach to give all terms in a document a context weight based on the contexts domain specific data. We have created a means of including our context weights in document ranking, by combining the weights with existing ranking models. Combining context weights with existing models has given us document-level term boosting, where the context of the queried terms within a document will positively or negatively affect the documents ranking score. We have tested out our approach by implementing a full search engine prototype and evaluatied it on a document collection within biomedical domain. Our work shows that this type of score boosting has little effect on overall retrieval precision. We conclude that the approach we have created, as implemented in our prototype, not to necessarily be good means of increasing precision in biomedical retrieval systems.
| Identifer | oai:union.ndltd.org:UPSALLA1/oai:DiVA.org:ntnu-8981 |
| Date | January 2009 |
| Creators | Johannsson, Dagur Valberg |
| Publisher | Norges teknisk-naturvitenskapelige universitet, Institutt for datateknikk og informasjonsvitenskap, Institutt for datateknikk og informasjonsvitenskap |
| Source Sets | DiVA Archive at Upsalla University |
| Language | English |
| Detected Language | English |
| Type | Student thesis, info:eu-repo/semantics/bachelorThesis, text |
| Format | application/pdf |
| Rights | info:eu-repo/semantics/openAccess |
Page generated in 0.0091 seconds