Skip to contents

birddog: sniffing emergence and trajectories in academic papers and patents

Source: R/birddog-package.R
birddog-package.Rd

Tools to detect emergence and trace technological/scientific trajectories in papers and patents. It reads OpenAlex and Web of Science data, builds citation-based networks, identifies groups, and summarizes their dynamics.

Pipeline

read_openalex()/read_wos() -> sniff_network() -> sniff_components() -> sniff_groups_cumulative() -> sniff_groups_lineage() -> sniff_trajectory_dag() -> sniff_trajectory_braid() -> analysis (sniff_trajectory_confluence(), _formation(), _destination(), _contribution(), _self_sufficiency(), _cct(), _entropy(), _hubs(), _dynamics()), each paired with a plot_* function. Focus an analysis with subset(flow, ...). "Flow" names the object kind, not an algorithm (stock vs flow: groups are stock, trajectories are flow): every detector returns a flow, and alternative detectors (sniff_trajectory_<algo>()) return the same contract, checked by validate_flow().

Label grammar

cNgN is a group (component N, group N); y<YYYY><cNgN> is a group-year node; tr::<cNgN> is a central trajectory (reaches the last year, one per final group); tr1..trN are absorbed trajectories. This grammar is part of the public API and only changes at major versions.

Theoretical background

Trajectories are detected as system-level objects (Dosi, 1982): disjoint chains in the temporal DAG of cumulative clusterings. A trajectory that stops being detected loses its identity, not its papers; tracking its terminal cohort to the last year classifies the outcome as convergence (one destination), divergence (many), or dormancy (none) — with emergence indicators on the living chains (Rotolo et al., 2015; Carley et al., 2017).

See also

Useful links:

Author

Maintainer: Roney Fraga Souza [email protected] (ORCID) [copyright holder]

Other contributors: