Author (Google Scholar) rankings dashboard

The simulation has been done on synthetic dataset. This visualizes the rankings of authors based on a composite score derived from their publication and citation data. The dashboard allows users to interactively explore a network of authors and view detailed metrics for each author.

How It Works

1. Data Loading:

   • The data is loaded from a JSON file (authors.json), which contains information about authors, including their citations, publications, and collaborators.

2. Composite Score Calculation:

   • Each author's ranking is computed based on a composite score that considers:
     • Total citations
     • Self-citation rate
     • Collaboration dependency
   • The formula used to calculate the composite score is:

     (citations * 0.4) - (self_citation_rate * 20) - (collaborator_dependency * 10)

   • The authors are sorted by this score, with higher scores indicating higher ranks.

3. Network Visualization:

   • A directed graph is created using NetworkX, where each author is represented as a node. The graph layout is generated using the spring_layout algorithm, which simulates a force-directed graph.
   • Authors are connected based on collaboration patterns, and the nodes are sized and colored based on their rank.

4. Interactive Dash Application:

   • The dashboard is built using Dash and Plotly. It features an interactive network graph where users can click on any author node to view more details about that author.
   • When a user clicks on a node, the following author metrics are displayed:
     • Rank
     • Composite Score
     • Citations
     • Publications
     • Self-citation Rate

Requirements

• dash
• plotly
• networkx
• json

You can install the required dependencies using pip:

pip install dash plotly networkx

Usage

from author_ranking import load_authors_data, generate_rankings, create_author_graph, generate_graph_figure

# Load data
authors_data = load_authors_data('path_to/authors.json')

# Generate rankings
rankings = generate_rankings(authors_data)

# Create a graph
G = create_author_graph(rankings)

# Generate a graph figure
fig = generate_graph_figure(G)