conexp-clj

This is conexp-clj, a general purpose software tool for Formal Concept Analysis.

The project has been started by Daniel Borchmann under supervision of Christian Meschke as part of the DFG project GA 216/10-1.

Features

conexp-clj is a pocket-calculator for Formal Concept Analysis. Its main purpose is to enable nontrivial examples to be computed easily.

conexp-clj features include:

• Basic Operations on Formal Contexts
• Relational Algebra with Formal Contexts
• Transparent IO for Formal Contexts (in development)
• Scaling for Many-Valued Contexts
• Implicational Theory and Basic Attribute Exploration
• NextClosure (of course)
• Computing Luxenburger-Bases and Iceberg Concept Sets
• IO for Many-Valued Contexts
• Lattice Layouts and Lattice IO (some...)
• A bit of Fuzzy-FCA
• Interface for sage

Note that conexp-clj is not a high-performance tool for Formal Concept Analysis. If you want this, check out Uta Priss' website on FCA software.

Prerequisites

You need

• a Jave Runtime Environment (≥ 1.6)
• Leiningen (≥ 2.0.0) if you want to run conexp-clj from source

How to Run

The recommended way to run conexp-clj is to download a pre-compiled version.

You can then run it through

$ java -jar conexp-clj-2.0.0-RC1.jar

from you command line (without the "$") to get a bare conexp-clj repl. If you want to try the experimental GUI, you can use

$ java -jar conexp-clj-2.0.0-RC1.jar -g

instead.

Documentation

The current main source of documentation on conexp-clj is its Wiki. Additionally, for general help on a function f, you can use the clojure function doc with

(doc f)

For finding functions you may find useful, you can use find-doc

(find-doc "Whatever you may find useful")

Additional Documentation:

• Basic example files: those files cover

  • basics
  • formal contexts
  • lattices
  • IO
  • implications

  They have been written by Sebastian Böhm.

• Advances example files:

  • Attribute Exploration a demonstration how attribute exploration can be done in conexp-clj.
  • Fuzzy FCA, a sample file to show how to use fuzzy FCA with conexp-clj
  • Factor Analysis, a small program that demonstrates how to use conexp-clj for factorizing contexts
  • Formal Contexts for Implications, computing a context for a set of implications
  • A Formal Context of Functions, see the paper by Artem Revenko and Sergej Kuznetzov for the CLA2010
  • Permutations as Formal Context, computes a context whose concept lattice is isomorphic to the lattice of permutations on the set {0,...,n}.
  • Tamari Lattice, the lattice of all bracketings of n+1 symbols (a.k.a. the Tamari Lattice of parameter n)

• A Tutorial, given at the 11th International Conference on Formal Concept Analysis

Running conexp-clj from source

To run conexp-clj from source, switch in the source directory of conexp-clj and run

$ lein deps

This will download any missing jar files needed for conexp-clj to run. To quickly obtain a repl just issue

$ lein repl

If you want a more sophisticated repl, you may try nrepl.el.

Compilation Instructions for conexp-clj

To create a standalone zip on your own, just run make in the top source directory. This will (should) do everything to create a zip archive containing a compiled version of conexp-clj. You can also run this compiled version directly by invoking ./conexp-clj/bin/conexp-clj (after unpacking the .zip file.)

Contributing Authors

See AUTHORS.md.

License

Copyright ⓒ 2009—2018 Daniel Borchmann

Distributed under the Eclipse Public License.

This program uses an adapted version of the G library, a 2D graphics library and rendering engine for Java, ⓒ 2009 GeoSoft, licensed under the GNU Lesser General Public License (LGPL). Modifications to the original version of G are only concerned with exposing internals necessary for conexp-clj to work. The modified version of G is again licensed under LGPL.

This program uses parts of the LatDraw library, ⓒ 2002 Ralph Freese.