DSRS: Estimation and Forecasting of Journal Influence in the Science and Technology Domain via a Lightweight Quantitative Approach

TL;DR

This paper introduces a lightweight, regression-based method for estimating journal influence scores without extensive data storage, validated against existing ranking systems, offering a practical alternative for assessing scientific impact.

Contribution

The paper presents a novel, data-efficient approach to journal influence estimation that reduces computational overhead compared to traditional citation-based methods.

Findings

The proposed method achieves low error compared to SCImago Journal Rank.

It does not require storing extensive citation data.

Validation shows the approach is feasible and effective.

Abstract

The evaluation of journals based on their influence is of interest for numerous reasons. Various methods of computing a score have been proposed for measuring the scientific influence of scholarly journals. Typically the computation of any of these scores involves compiling the citation information pertaining to the journal under consideration. This involves significant overhead since the article citation information of not only the journal under consideration but also that of other journals for the recent few years need to be stored. Our work is motivated by the idea of developing a computationally lightweight approach that does not require any data storage, yet yields a score which is useful for measuring the importance of journals. In this paper, a regression analysis based method is proposed to calculate Journal Influence Score. Proposed model is validated using historical data from the SCImago portal. The results show that the error is small between rankings obtained using the proposed method and the SCImago Journal Rank, thus proving that the proposed approach is a feasible and effective method of calculating scientific impact of journals.