A New Metric for the Analysis of the Scientific Article Citation Network

Livia Lin-Hsuan Chang,Frederick Kin Hing Phoa,Junji Nakano

doi:10.1109/access.2019.2937220

Abstract

Citation plays an important role in the bibliometrics analysis since the introduction of the impact factors, but traditional measures mainly focused on the direct citations between articles. In this work, we introduce a new metric, namely Article Network Influence (ANI), to measure the influence of an article by using broader citation relationships quantitatively. We prepare our article citation networks from one of the largest citation databases called the Web of Science, and we demonstrate the use of ANI on the analysis of these networks in the statistics research community. These analyses appear in the top-20 influential articles in statistics within every 11 years during 1981-2016. We consider differences between the new metric and several traditional measures, including the impact factor, PageRank, and Field-Weighted Citation Impact (FWCI).

Highlights

Bibliometrics [1] is an analysis of research performance that governments and research institutes advocate and emphasize nowadays
Following the spirit of the impact factor, we extend the use of Article Network Influence (ANI) to the journal-based level via considering the location of ANI of the articles published in the journal, namely Journal Network Influence (JNI)
When one considers an aggregated ANI from several articles, Field-Weighted Citation Impact (FWCI) takes the arithmetic means on the ratio of actual to the expected citations of each publication while JNI defined in section IV-A takes the median values of ANI of each publication

Summary

A New Metric for the Analysis of the Scientific Article Citation Network

This work was supported in part by the Career Development Award of Academia Sinica, Taiwan, under Grant 103-CDA-M04, in part by the Ministry of Science and Technology, Taiwan, under Grant 107-2118-M-001-011-MY3 and Grant 107-2321-B-001-038, in part by the ISM Cooperative Research Program under Grant 2016-ISMCRP-4408, Grant 2017-ISMCRP-4206, and Grant 2017-ISMCRP-4210, and in part by SOKENDAI Publication Grant for Research Papers.

INTRODUCTION

A COMPARISON TO PAGERANK ALGORITHM IN

CONCLUSION