You are here
Automatic algorithm selection for MeSH Heading indexing based on meta-learning.
We present a methodology that automatically selects indexing algorithms for each heading in MeSH, NLM’s vocabulary for indexing MEDLINE.While manually comparing indexing methods is manageable with a limited number of MeSH headings, a large number of them makes automation of this selection desirable. Results show that this process can be automated based on previously indexed MEDLINE records. We find that AdaBoostM1 is better suited to index a group of MeSH headings named Check Tags and helps improve the micro F-measure from 0.5385 to 0.7157 and the macro F-measure from 0.4123 to 0.5387 (both p < 0.01).