High acknowledgement index: Characterizing research supporters with factors of acknowledgement affecting paper citation counts.

• This study proposes an automated acknowledgement index and explores its relationship with researchers' performance. • We performed sensitivity analysis to understand the impact of acknowledgement types, automatically classified through NLP techniques, on paper citation counts. • The study presents a method for identifying research supporters amidst challenges posed by similar names and affiliations. The acknowledgments section of scientific papers is paramount in academic production, providing valuable insights into the individuals and organizations involved in the research process. Developing a standard index of acknowledgments would provide a significant metric for measuring institutional and researcher influence. However, the absence of standardized formats has impeded effective analysis of information in acknowledgement sections. This study develop an acknowledgement index utilizing automated text-mining techniques to address these limitations. In addition, we propose two methods to disambiguate research supporters in the acknowledgement section: entity scores based on specific keywords and research similarity scores calculated through co-references. Based on our investigation, we explore the characteristics of research supporters with high acknowledgement index scores and examine the correlation between each acknowledgement index and their performance. Notably, paper citation strongly correlates with research supporters' performance. Next, our analysis delves into the impact of acknowledgement and entity types on paper citations. Our findings reveal that acknowledgments with only person names exert the most significant impact on paper citation. [ABSTRACT FROM AUTHOR]