It should be noted that the citation-based method (Amsler) has a very low precision, recall, and F1 measure, which means that the relation between the citation-based method based on the Amsler network and the benchmark considering the text-based similarity based on TF-IDF is weak; therefore, the difference between the citation-based and text-based clustering methods is highly significant.
← all excerpts
Hybrid self-optimized clustering model based on citation links and textual features to detect research topics.
1
—
—