TY - GEN
T1 - Ontology-Based Semantic Similarity Approach for Biomedical Dataset Retrieval
AU - Wang, Xu
AU - Huang, Zhisheng
AU - van Harmelen, Frank
PY - 2020
Y1 - 2020
N2 - Ontology-based semantic similarity approaches play an important role in text-similarity task, thanks to its ability of explanation. Ontology-based semantic similarity approaches can explain how two terms are similar with help of rich knowledge in ontology. Information retrieval aims to find relevant information for given user query. As a subareas of information retrieval, dataset retrieval is an activity to find dataset which are relevant to an information need, by using full-text indexing approach or content-based indexing approach. Ontology-based semantic similarity approaches can not only do some information retrieval tasks, such as full-text mapping, but also finding deeper similar information with the help of knowledge-richness in ontology. Because of the advantage of ontology-based similarity approaches, we are looking forwards to find the possibility to using ontology-based similarity for datasets retrieval. In this paper, we provide an ontology-based similarity approach for dataset retrieval. We run our novel approach on the bioCADDIE 2016 Dataset Retrieval Challenge. After ruining experiments, we evaluate our results with several information retrieval evaluation measures. The evaluation results show that our approach could perform well.
AB - Ontology-based semantic similarity approaches play an important role in text-similarity task, thanks to its ability of explanation. Ontology-based semantic similarity approaches can explain how two terms are similar with help of rich knowledge in ontology. Information retrieval aims to find relevant information for given user query. As a subareas of information retrieval, dataset retrieval is an activity to find dataset which are relevant to an information need, by using full-text indexing approach or content-based indexing approach. Ontology-based semantic similarity approaches can not only do some information retrieval tasks, such as full-text mapping, but also finding deeper similar information with the help of knowledge-richness in ontology. Because of the advantage of ontology-based similarity approaches, we are looking forwards to find the possibility to using ontology-based similarity for datasets retrieval. In this paper, we provide an ontology-based similarity approach for dataset retrieval. We run our novel approach on the bioCADDIE 2016 Dataset Retrieval Challenge. After ruining experiments, we evaluate our results with several information retrieval evaluation measures. The evaluation results show that our approach could perform well.
KW - Biomedical dataset
KW - Dataset retrieval
KW - Semantic similarity
UR - http://www.scopus.com/inward/record.url?scp=85094144162&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85094144162&partnerID=8YFLogxK
U2 - 10.1007/978-3-030-61951-0_5
DO - 10.1007/978-3-030-61951-0_5
M3 - Conference contribution
SN - 9783030619503
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 49
EP - 60
BT - Health Information Science
A2 - Huang, Zhisheng
A2 - Siuly, Siuly
A2 - Wang, Hua
A2 - Zhang, Yanchun
A2 - Zhou, Rui
PB - Springer Science and Business Media Deutschland GmbH
T2 - 9th International Conference on Health Information Science, HIS 2020
Y2 - 20 October 2020 through 23 October 2020
ER -