Replication Data for: Revealing the shifts in sustainable mining and critical mineral strategies: A multi-method text mining approach in Türkiye (1963–2028)

Description

This repository contains the comprehensive raw textual dataset, custom Python source code, and supplementary materials supporting the manuscript "Revealing the shifts in sustainable mining and critical mineral strategies: A multi-method text mining approach in Türkiye (1963–2028)." The dataset comprises 65 years (1963–2028) of national planning documents, including 12 National Development Plans and 14 Special Expert Commission Reports (SECRs) on mining, coal, and metal minerals. The provided Python script allows for the full reproduction of the multi-method NLP pipeline, including regex-based preprocessing, TF-IDF signature keyword extraction, Latent Dirichlet Allocation (LDA) topic modeling, semantic co-occurrence network matrix generation, and lexicon-based risk sentiment analysis. These materials, along with the domain-specific stop-word lists, are shared to promote strict transparency and computational reproducibility in mineral governance and policy research.

Steps to reproduce

Download and extract the repository files to your local machine. Ensure that Python 3.x is installed on your system. Open your terminal or command prompt and install the required dependencies by running: pip install PyMuPDF pandas scikit-learn networkx numpy openpyxl Ensure that the "Python_NLP_Pipeline_Code.py" script and the "Supplementary_Table_S1_StopWords.xlsx" file are in the same root directory alongside the "/corpus/" folder. Execute the "Python_NLP_Pipeline_Code.py" script. The script will automatically process the PDFs, apply the domain-specific stop-words, and generate an "/output/" folder containing the 1000 TF-IDF keywords and the Nodes/Edges CSV files necessary for VOSviewer network mapping.

Institutions

• Pamukkale University

  Denizli, Denizli

Categories

Licence