💭 RubyFCA
Command line tool for Formal Concept Analysis (FCA) written in Ruby.
Change Log
- Microsoft Excel
.xlsxfiles supported [June 18, 2023] - Sample files added [June 18, 2023]
Features
- Converts data in XLSX (Microsoft Excel), CSV (comma-separated values), or CXT (Conexp and generate a Graphviz DOT file, a SVG/EPS vector image file, or /a PNG/JPG bitmap file.
- Adopts the Ganter algorithm (with reference to its Perl implementation of Fcastone by Uta Priss).
Dependencies
For example, to install Graphviz using Homebrew on MacOS, execute the following command
brew install graphviz
Installation
Install the gem:
gem install rubyfca
How to Use
RubuFCA converts Conexp CXT data to Graphviz dot format.
Usage:
rubyfca [] <source file> <output file>
where:
<source file>
".xlsx", ".csv" ,".cxt"
<output file>
."svg", ".png", ".jpg", ".eps", or ".dot"
[]:
--full, -f: Do not contract concept labels
--coloring, -c <i>: Color concept nodes [0 = none (default), 1 =
lightblue/pink, 2 = monochrome] (default: 0)
--straight, -s: Straighten edges (available when output format is
either png, jpg, svg, pdf, or eps)
--nodesep, -n <f>: Size of separation between sister nodes (from 0.1 to
5.0) (default: 0.4)
--ranksep, -r <f>: Size of separation between ranks (from 0.1 to 5.0)
(default: 0.2)
--legend, -l: Print the legend of concept nodes (available only when
using circle node shape)
--circle, -i: Use circle shaped concept nodes
--version, -v: Print version and exit
--help, -h: Show this
Examples
Input Data
XLSX (Excel)
CSV
, Ostrich , Sparrow , Eagle , Lion , Bonobo , Human being
bird , X , X , X , , ,
mammal , , , , X , X , X
ape , , , , , X , X
, , X , X , , ,
, , , X , X , ,
talking , , , , , , X
CXT
B
6
6
Ostrich
Sparrow
Eagle
Lion
Bonobo
Human being
bird
mammal
ape
talking
XXX...
...XXX
....XX
.XX...
..XX..
.....X
Output
rubyfca input_file output_file --coloring 1 --full --nodesep 0.8 --ranksep 0.3 --straight
rubyfca input_file output_file --coloring 2 --nodesep 0.5 --ranksep 0.3
rubyfca input_file output_file --circle --legend --coloring 1 --full --nodesep 0.8 --ranksep 0.3 --straight
Copyright
Copyright (c) 2009-2023 Yoichiro Hasebe and Kow Kuroda. See LICENSE for details.