A New Journal Ranking Method: The Reputation Analysis of Citation Behavior Model

Abstract

The authority of journals is usually a key indicator in guiding people to choose important journals, and the journal ranking is a common way to distinguish the journals' authority. The common ranking methods or models are based on citation data, such as the impact factor (IF), the PageRank algorithm, and the hyperlink-induced topic search algorithm. In this paper, we present a new model, named the reputation analysis of citation behavior (RACB) model, which not only considers the number of citations but also considers the reputation of inter-citation behavior. First, the model fits the function relation of the citation desire index (CDI) based on the citation data of the most reputable journal (named the top journal in the remainder of this paper). Then, the CDI values of the target journals are calculated by using the function relation of the top journal. An improved gray correlation analysis is used to describe the deviation between the CDI and the actual citation rate, named the random citation rate, indicating the inter-citation reputation value of the target journals. Finally, a case study showed that the ranking result of the RACB model has a high similarity with the IF values of 2017, which indicates that the ranking results of the RACB model that considered journal reputation evaluations are more reasonable than those of the traditional PageRank algorithm. This paper proposes a new method of journal ranking from the perspective of reputation evaluation, which can rank the journals more reasonably. It encourages researchers to cite articles more fairly in order to avoid a situation in which the low reputation of the citation behavior affects the quality of the journal that published the article.

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 — 7 citations exist upstream.

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