Semantic Relationship Study between Citing and Cited Scientific Articles Using Topic Modeling

Abstract

In the evaluation of scientific research, it is clear that 'scientific quality' is proving to be one of the most important criteria, and in the case of research for development, it is on an equal footing with another key dimension: relevance for development. The evaluation criteria and procedures are very similar to those applied in the evaluation of research in general: peer reviews, panels, bibliometric indices, etc ... The practical application of these tools in the case of research for development can sometimes be quite difficult. There is, however, a general consensus that quality should be expressed identically for any scientific research, with the same problems and constraints. In this article and to evaluate a scientific article, we thought to study the semantic relationship between the citing article and the other cited, we based on the Topic modeling theme with the choice of LDA and LDA2VEC algorithms applied to the corpus Nips (1987-2017) then on the citations of each article of the corpus.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 6 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.