Citation Network Analysis from Scratch
summary
Step-by-step recipe for reproducing a citation network: pull reference lists from Scopus via its API or open Crossref dumps (Crossref open citations grows from 1% to ~88% coverage), build a raw adjacency matrix of citing → cited pairs, convert it to a co‑citation or bibliographic coupling matrix via transpose × original and apply cosine similarity, then run principal component analysis or factor analysis and use NetworkX/igraph for graphs and visualization; code and preprint are promised on GitHub, but citations measure influence not paper quality.
AI-written summary · about
chapters (5)
- 0:00Start
- 0:58Citation Network Analysis (from scratch)
- 3:21Delineation of the Research Field (keyword searches in a database)
- 9:09Construction of the Network (and calculating connectedness measure)
- 12:30Multivariate Statistical Analysis Principal component analysis
related talks
01h 12mSocial Network Analysis in Python
01h 06mComplex networks sparsification: methods and tools
01h 11mUsing Molecular Networks to Seed Predictive Hierarchies of Cell Systems
37mPageRank all the things!
21mUnleashing the Power of Data Science
44mPython for Economists: An overview of Python tools for Economists
35mTowards Interpretable Data Science
34mWhat will a whole nation do? Predicting events in human networks