A Neural Learning Approach for Prediction of Research Citations Using Article Semantics

Abstract

Predicting scholarly output largely depend on the research citations received right from the inception. Classic research articles across multiple research domains as identified by Google Scholar tend to define articles published in 2006 which receive citations even after 10 years and counting. Though author and journal level features tend to attract early stage citations, it is for sure that only the righteousness in the article content would have triggered the continuity of citations all through the rest of years. Thus it is clear that for receiving `citation sustainability', an article has to be well conceived, constructed and delivered. This work presents a neural learning approach that uses article semantic features for predicting the intensity of research citations for classic research articles.

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 — 5 citations exist upstream.

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