Parsimonious citer-based measures: The artificial intelligence domain as a case study

This article presents a new Parsimonious Citer-Based Measure for assessing the quality of academic papers. This new measure is parsimonious as it looks for the smallest set of citing authors (citers) who have read a certain paper. The Parsimonious Citer-Based Measure aims to address potential distortion in the values of existing citer-based measures. These distortions occur because of various factors, such as the practice of hyperauthorship. This new measure is empirically compared with existing measures, such as the number of citers and the number of citations in the field of artificial intelligence (AI). The results show that the new measure is highly correlated with those two measures. However, the new measure is more robust against citation manipulations and better differentiates between prominent and nonprominent AI researchers than the above-mentioned measures.