Genealogical Trees of Scientific Papers

Area of Science:

• Bibliometrics
• Network Science
• Information Science

Background:

• Scientific citation databases are often studied using network analysis.
• Predicting paper popularity and ranking articles by in-degree are established methods.
• The temporal dynamics and evolution of citation networks are emerging research areas.

Purpose of the Study:

• To propose an original algorithm for constructing genealogical trees of scientific papers.
• To analyze the evolutionary perspective of scientific paper citation networks.
• To identify research milestones through citation count evolution.

Main Methods:

• Developing an algorithm based on the time evolution of citation counts.
• Defining paper 'fitness' by its in-degree growing trend.
• Observing the decline in trend for 'dying' papers and the rise of 'offspring' papers.

Main Results:

• The algorithm successfully constructs genealogical trees based on citation dynamics.
• Paper fitness is effectively measured by in-degree growth trends.
• The model captures the 'birth' and 'takeover' dynamics of scientific papers.

Conclusions:

• This evolutionary approach provides a new method for understanding scientific impact over time.
• Genealogical trees can visualize the lineage and evolution of research fields.
• The algorithm can be practically applied to trace key publications and research milestones.