Limits...
Integrating information retrieval with distant supervision for gene ontology annotation.

Zhu D, Li D, Carterette B, Liu H - Database (Oxford) (2014)

Bottom Line: Then, a greedy approach was applied to associate genes with sentences.Our best performing system for subtask A achieves an F1 score of 0.27 based on exact match and 0.387 allowing relaxed overlap match.Our best performing system for subtask B, a search-based system, achieves an F1 score of 0.075 based on exact match and 0.301 considering hierarchical matches.

View Article: PubMed Central - PubMed

Affiliation: Department of Health Sciences Research, Mayo Clinic, 200 First St SW, Rochester, MN 55905 and Department of Computer & Information Sciences, University of Delaware, 101 SMITH HALL, Newark, DE 19716, USA Department of Health Sciences Research, Mayo Clinic, 200 First St SW, Rochester, MN 55905 and Department of Computer & Information Sciences, University of Delaware, 101 SMITH HALL, Newark, DE 19716, USA.

Show MeSH

Related in: MedlinePlus

Overview of System B2.
© Copyright Policy - creative-commons
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4150992&req=5

bau087-F2: Overview of System B2.

Mentions: Figure 2 gives an overview of System B2, which has similar modules to System B1. The major difference is that we use GeneRIF (6) as the external resource. In particular, we extract <Sentence, GOID> pairs from GeneRIF where the corresponding articles are cited as evidence of GOA records in iProClass and built an index for this collection of sentences. Therefore, the output from the Retrieval model is a ranked list of sentences, which we further converted to a ranked list of GOID based on <Sentence, GOID> pairs. Finally, in the Annotation module, we did the following:


Integrating information retrieval with distant supervision for gene ontology annotation.

Zhu D, Li D, Carterette B, Liu H - Database (Oxford) (2014)

Overview of System B2.
© Copyright Policy - creative-commons
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4150992&req=5

bau087-F2: Overview of System B2.
Mentions: Figure 2 gives an overview of System B2, which has similar modules to System B1. The major difference is that we use GeneRIF (6) as the external resource. In particular, we extract <Sentence, GOID> pairs from GeneRIF where the corresponding articles are cited as evidence of GOA records in iProClass and built an index for this collection of sentences. Therefore, the output from the Retrieval model is a ranked list of sentences, which we further converted to a ranked list of GOID based on <Sentence, GOID> pairs. Finally, in the Annotation module, we did the following:

Bottom Line: Then, a greedy approach was applied to associate genes with sentences.Our best performing system for subtask A achieves an F1 score of 0.27 based on exact match and 0.387 allowing relaxed overlap match.Our best performing system for subtask B, a search-based system, achieves an F1 score of 0.075 based on exact match and 0.301 considering hierarchical matches.

View Article: PubMed Central - PubMed

Affiliation: Department of Health Sciences Research, Mayo Clinic, 200 First St SW, Rochester, MN 55905 and Department of Computer & Information Sciences, University of Delaware, 101 SMITH HALL, Newark, DE 19716, USA Department of Health Sciences Research, Mayo Clinic, 200 First St SW, Rochester, MN 55905 and Department of Computer & Information Sciences, University of Delaware, 101 SMITH HALL, Newark, DE 19716, USA.

Show MeSH
Related in: MedlinePlus