Abstract
Original language | English |
---|---|
Journal | Multimedia Tools and Applications |
Volume | 71 |
Issue number | 2 |
Pages (from-to) | 905-929 |
Number of pages | 25 |
ISSN | 1380-7501 |
DOIs | |
Publication status | Published - Jul 2014 |
Fingerprint
Cite this
}
Expanding user's query with tag-neighbors for effective medical information retrieval. / Durao, F.; Bayyapu, Karunakar; Xu, Guandong; Dolog, P.; Lage, Ricardo Gomes.
In: Multimedia Tools and Applications, Vol. 71, No. 2, 07.2014, p. 905-929.Research output: Contribution to journal › Journal article › Research › peer-review
TY - JOUR
T1 - Expanding user's query with tag-neighbors for effective medical information retrieval
AU - Durao, F.
AU - Bayyapu, Karunakar
AU - Xu, Guandong
AU - Dolog, P.
AU - Lage, Ricardo Gomes
PY - 2014/7
Y1 - 2014/7
N2 - Medical information is a natural human demand. Existing search engines on the Web often are unable to handle medical search well because they do not consider its special requirements. Often a medical information searcher is uncertain about his exact questions and unfamiliar with medical terminology. Under-specified queries often lead to undesirable search results that do not contain the information needed. To overcome the limitations of under-specified queries, we utilize tags to enhance information retrieval capabilities by expanding users' original queries with context-relevant information. We compute a set of significant tag neighbor candidates based on the neighbor frequency and weight, and utilize the qualified tag neighbors to expand an entry query. The proposed approach is evaluated by using MedWorm medical article collection and results show considerable precision improvements over state-of-the-art approaches.
AB - Medical information is a natural human demand. Existing search engines on the Web often are unable to handle medical search well because they do not consider its special requirements. Often a medical information searcher is uncertain about his exact questions and unfamiliar with medical terminology. Under-specified queries often lead to undesirable search results that do not contain the information needed. To overcome the limitations of under-specified queries, we utilize tags to enhance information retrieval capabilities by expanding users' original queries with context-relevant information. We compute a set of significant tag neighbor candidates based on the neighbor frequency and weight, and utilize the qualified tag neighbors to expand an entry query. The proposed approach is evaluated by using MedWorm medical article collection and results show considerable precision improvements over state-of-the-art approaches.
UR - http://www.scopus.com/inward/record.url?scp=84871438600&partnerID=8YFLogxK
U2 - 10.1007/s11042-012-1316-5
DO - 10.1007/s11042-012-1316-5
M3 - Journal article
VL - 71
SP - 905
EP - 929
JO - Multimedia Tools and Applications
JF - Multimedia Tools and Applications
SN - 1380-7501
IS - 2
ER -