Predictable by publication: discovery of early highly cited academic papers based on their own features

Abstract

PurposePredicting highly cited papers can enable an evaluation of the potential of papers and the early detection and determination of academic achievement value. However, most highly cited paper prediction studies consider early citation information, so predicting highly cited papers by publication is challenging. Therefore, the authors propose a method for predicting early highly cited papers based on their own features.Design/methodology/approachThis research analyzed academic papers published in the Journal of the Association for Computing Machinery (ACM) from 2000 to 2013. Five types of features were extracted: paper features, journal features, author features, reference features and semantic features. Subsequently, the authors applied a deep neural network (DNN), support vector machine (SVM), decision tree (DT) and logistic regression (LGR), and they predicted highly cited papers 1–3 years after publication.FindingsExperimental results showed that early highly cited academic papers are predictable when they are first published. The authors’ prediction models showed considerable performance. This study further confirmed that the features of references and authors play an important role in predicting early highly cited papers. In addition, the proportion of high-quality journal references has a more significant impact on prediction.Originality/valueBased on the available information at the time of publication, this study proposed an effective early highly cited paper prediction model. This study facilitates the early discovery and realization of the value of scientific and technological achievements.

Editorial: Special selection on contemporary bibliometric analytics

“…Readers may be interested in some related articles on whether authors play fair or manipulate Google Scholar bibliometric indicators, such as h-index and i10-index (Loan and Shah, 2022), and the analysis of Sci-Hub use with Google Trends (Behboudi et al ., 2021); Chen et al . (2021) studied the effect of interdisciplinary components' citation intensity on scientific impact, whether proceedings papers in science fields have higher impacts than those in social science and humanities (Yang and Qi, 2021), characteristics analysis and evaluation of discourse leading for academic journals (Wang, 2022), the ecosystem of research tools for scholarly communication (Rao et al ., 2022), the scientific outcome in the domain of grey literature (Wani and Ganaie, 2022), early discovering highly cited academic papers (Tang et al ., 2023) and textbook citations (Maleki et al ., 2023).…”

A Study on Iconic Paper Discovery Based on Deep Learning Models

“…High citation papers are important evaluation indicators for research competitiveness, research performance, and hotspot analysis (Ma et al, 2008). However, the evaluation of research achievements based solely on highly cited papers is too limited to comprehensively assess the value and contribution of research (Tang et al, 2023). In recent years, many scholars have started to focus on the research of representative works, trying to supplement the evaluation of highly cited papers.…”