Application of a Novel Subject Classification Scheme for a Bibliographic Database Using a Data-Driven Correspondence.

Area of Science:

• Bibliometrics
• Information Science
• Data Science

Background:

• Research evaluation often requires updating subject classification schemes in bibliographic databases.
• Manual reclassification is labor-intensive and time-consuming, hindering efficient research assessment.
• Existing classification systems may not align with emerging research trends or specific evaluation needs.

Purpose of the Study:

• To develop an efficient and data-driven method for applying a novel subject classification scheme to a pre-classified bibliographic database.
• To establish a robust model for mapping between existing and new classification systems.
• To facilitate improved research evaluation through accurate and updated subject categorization.

Main Methods:

• A subject classification model based on topological spaces was defined for bibliographic databases.
• A data-driven correspondence between new and existing subject classification schemes was established using a research project database.
• The approach focused on forming a compact topological space for the novel classification scheme.

Main Results:

• The proposed approach successfully applied a novel subject classification to a practical research evaluation tool.
• The method demonstrated efficiency in mapping between two distinct subject classification schemes.
• The case study involved integrating a new classification into a proprietary citation database for benchmarking.

Conclusions:

• The data-driven, topological space-based approach offers an effective solution for updating subject classifications in bibliographic databases.
• This method significantly reduces the labor and time associated with reclassification for research evaluation.
• The approach is applicable to real-world scenarios, enhancing the utility of citation databases for benchmarking and assessment.