Skip to content

Latest commit

 

History

History
43 lines (32 loc) · 972 Bytes

README.md

File metadata and controls

43 lines (32 loc) · 972 Bytes

kmapper_law_analysis

Korean law data analysis using Topological Data Analysis and Mapper Algorithm


Installation & Execution

Prerequistes

  • Python (>= 3.6)
  • NumPy
  • Scikit-learn
  • openpyxl

1. Clone

$ git clone https://github.com/zeebraa00/kmapper_law_analysis.git

2. Install Packages

$ pip install numpy
$ pip install -U scikit-learn
$ pip install openpyxl

3. Make custom distance matrix of data

$ python make_metric.py

We have focused on the reference relation of law.

  1. Initiate the distance matrix. (Set the distance between all laws to 1.)
  2. Shorten distance between laws used in same precedent while scanning Korean precedents.
  3. The completed distance matrix is saved as a binary file. (law_data/custom_metric.npy)

4. Cluster Korean-law-data using kepler-mapper

$ python main.py

Copyright ⓒ 2021 성균관대학교 수학과 김성훈, 정재헌 All Rights Reserved