Where Do Your Citations Come From? Citation-Constellation: A Free, Open-Source, No-Code, and Auditable Tool for Citation Network Decomposition with Complementary BARON and HEROCON Scores
In recent years, the academic community has raised concerns regarding the limitations of standard citation metrics, which treat all citations as equal. This approach often obscures the complex social and structural pathways through which scholarly influence propagates. To address these issues, a new tool named Citation-Constellation has been developed, providing a comprehensive analysis of citation networks.
According to the paper published on arXiv (arXiv:2603.24216v1), Citation-Constellation is a freely available, no-code tool designed for citation network analysis. It introduces two complementary bibliometric scores: BARON (Boundary-Anchored Research Outreach Network score) and HEROCON (Holistic Equilibrated Research Outreach CONstellation score). These scores allow for a detailed decomposition of a researcher’s citation profile based on the network proximity between citing and cited authors.
Understanding BARON and HEROCON Scores
BARON is a strict binary metric that counts only citations originating from outside the detected collaborative network. In contrast, HEROCON applies graduated weights that assign partial credit to in-group citations based on relationship proximity. The difference between these two scores serves as a diagnostic tool to measure inner-circle dependence within a researcher’s citation network.
Phased Architecture of Citation-Constellation
The Citation-Constellation tool implements its functionality through a phased architecture, which includes:
- Self-citation analysis: Evaluating the extent to which researchers cite their own work.
- Co-authorship graph traversal: Analyzing collaboration patterns among co-authors.
- Temporal institutional affiliation matching via ROR: Aligning citations with the institutions researchers were affiliated with at different times.
- AI-agent-driven venue governance extraction: Utilizing a local large language model (LLM) to extract governance information about citation venues (currently under development).
Key Design Features
Several key design choices have been made in the development of Citation-Constellation:
- ORCID-validated author identity resolution: Ensuring accurate author identification through the use of ORCID IDs.
- UNKNOWN classification: Handling citations with insufficient metadata by categorizing them as UNKNOWN.
- Comprehensive audit trails: Maintaining detailed documentation of every classification decision made within the tool.
- No-code web interface: Enabling researchers to compute scores without the need for programming, installation, or registration.
Conclusion and Future Directions
It is important to note that the BARON and HEROCON scores are presented as structural diagnostics rather than quality indicators. They provide insights into the social graph from which citations originate and should not be used for hiring, promotion, or funding decisions. The HEROCON weights are still experimental and will require further empirical calibration.
As the academic community continues to seek more nuanced insights into citation practices and scholarly influence, tools like Citation-Constellation represent a significant step forward in citation network analysis.