fcaR 1.5.0

Improvements:

• OOP Refactoring: ImplicationSet now inherits from RuleSet, following proper object-oriented design. Shared methods (filtering, subsetting, printing, serialization, etc.) live in the parent RuleSet class, eliminating ~600 lines of duplicated code. ImplicationSet retains only FCA-specific methods (closure(), apply_rules(), to_basis(), to_direct_optimal(), etc.).
• Tidyverse for RuleSet: The dplyr verbs (filter(), arrange(), slice()) now work on RuleSet objects in addition to ImplicationSet, via S3 method dispatch.

fcaR 1.4.1

• JSON Import/Export: Added to_json() methods and corresponding *_from_json() functions for FormalContext, ConceptLattice, ImplicationSet, and RuleSet. This allows for efficient serialization of all major data structures in fcaR, including recursive export of nested objects.

fcaR 1.4.0

Major Enhancements:

• Tidyverse integration: Implemented S3 methods to support dplyr verbs, allowing for a fluent, grammar-based manipulation of FCA objects:
  • FormalContext: Support for select() (attributes), filter() (objects), mutate() (feature engineering), arrange() (sorting), and rename(). Includes support for tidyselect helpers (e.g., starts_with()).
  • ImplicationSet: Support for filter() (based on metrics or attributes), arrange() (sorting rules), and slice() (subsetting by index).
• Semantic Rule Filtering: Introduced helper functions for ImplicationSet filtering to query rules based on attribute presence/absence: lhs(), rhs(), not_lhs(), lhs_any(), etc. This allows querying rules like filter(rhs("Attribute_A"), support > 0.2).
• Mining Causal Association Rules: Implemented the find_causal_rules() method in FormalContext, enabling the discovery of causal rules by controlling for confounding variables using the “Fair Odds Ratio” on matched pairs.

Improvements:

• Robust Subsetting: Completely rewritten subcontext() method in FormalContext. It now robustly handles negative indices, logical vectors, and character vectors, and prevents dimension collapsing issues (using drop = FALSE) that previously caused errors with the Matrix package.
• Metadata Preservation: Rewritten [ and related methods of ImplicationSet. These ensure that critical context metadata (such as the number of objects \(N\) for support calculation) is preserved when filtering or sorting rules, fixing previous issues where metadata was lost.
• Data Safety: Enhanced type safety in internal functions to strictly handle integer indices, preventing errors with dplyr attributes.

Documentation:

• New Vignette: Added fcaR_dplyr vignette illustrating the new data manipulation workflow.
• New Vignette: Added causal vignette explaining the new causal mining functionality and its application to Simpson’s Paradox.

Fixes:

• Fixed Matrix coercion errors (dgCMatrix to data.frame) in R 4.x when using internal incidence matrices.
• Fixed fixupDN.if.valid errors from the Matrix package when filtering operations resulted in empty contexts (0 objects or 0 attributes).
• Resolved floating-point precision issues in unit tests when comparing support values.
• Reduced dependencies: moved ggplot2, ggraph, igraph, rstudioapi, and yaml to Suggests.
• Removed forcats and magrittr dependencies by using base R equivalents and the native pipe |>.

fcaR 1.3.1

• Added tests

fcaR 1.3.0

Major Enhancements:

• Matrix factorization: Added factorize() method to FormalContext class. It now implements two state-of-the-art algorithms:
  • GreConD+: For Boolean and Fuzzy matrix factorization with grades and overcovering (Belohlavek & Trneckova, 2024). It fully supports custom fuzzy logics set in the context.
  • ASSO: A heuristic algorithm based on association rules for Boolean matrix factorization.
• Advanced randomization: New suite of functions to generate and perturb datasets for statistical testing:
  • RandomContext(): Generates synthetic contexts using Uniform or Dirichlet distributions (mimicking real-world data structure).
  • randomize_context(): Randomizes existing contexts via Edge swapping (preserves marginal sums) or Rewiring (preserves density).
  • Distributive Generators: Added RandomDistributiveContext() to generate synthetic data guaranteed to produce distributive lattices (based on Birkhoff’s theorem).
• Robustness & performance: The InClose algorithm has been completely refactored to use the C-API directly (avoiding Rcpp overhead). This significantly improves performance for sparse matrices.

New Functionality:

• Advanced metrics: Added methods stability(), separation(), and fuzzy_density() to ConceptLattice to compute concept quality metrics.
• Lattice properties: Added methods to ConceptLattice to efficiently check algebraic properties using sparse matrix operations: is_distributive(), is_modular(), is_semimodular(), and is_atomic().
• Fuzzy algorithms: In fuzzy contexts, the user can now select the algorithm for find_concepts(): “InClose” (default), “FastCbO”, or “NextClosure”.
• Implications:
  • Added to_direct_optimal() to convert implication sets into the direct optimal basis.
  • Added use_hedge() and get_hedge() to manage hedges in fuzzy implication closures.
• Repo integration: The connection to fcarepository.org is now managed more elegantly, including an RStudio addin to fetch contexts.

Improvements:

• Visualization: Removed dependency on hasseDiagram. Implemented a new native graphics engine for concept lattices.
• Efficiency: Optimized C functions for concept support computation and general lattice mining. Also, the NextClosure implementation for building the Duquenne-Guigues basis of implications in the binary case has been optimized.
• Documentation: Added new vignettes: advanced_lattice_metrics, creating_contexts, fuzzy_fca, lattice_visualization, matrix_factorization, random_contexts and lattice_properties.

fcaR 1.2.2

Enhancements:

• Added more unit tests.
• Minor changes to the plotting of formal contexts.
• Now the fc$scale() function admits a new argument bg (default: FALSE) which, if set to TRUE, avoids computing the background knowledge of the scales.

Fixes:

• Changed the package documentation format to comply with the new roxygen2 _PACKAGE sentinel.

fcaR 1.2.1

Enhancements:

• Other logics have been implemented. Now, we can use fc$use_logic() to select one of the available_logics().
• Improved export to LaTeX.

Bugfixes:

• Fixes required by the new version of Matrix and the new use of HTML Tidy in R 4.2.
• Some rounding errors might induce errors in the computations. These has been fixed.

fcaR 1.2.0

• Fixes required by the new version of Matrix and the new use of HTML Tidy in R 4.2.

fcaR 1.1.1

Enhancements:

• The user can control the number of decimal digits when exporting to LaTeX or when printing formal contexts, concept lattices and implications. Just use fcaR_options(decimal_places = n), where n is the number of desired decimal digits.

New functionality:

• Now the package uses the settings package to manage several options. Currently, the only option is the number of decimal digits to use when printing or exporting to LaTeX.

Bugfixes:

• Fixed exporting to latex with special characters such as $, _, etc.

fcaR 1.1.0

Enhancements:

• Better printing of Sets.
• More optimized ConceptLattice class. Now it inherits from a ConceptSet superclass with generic functions.

New functionality:

• Added function to compute the difference of two Sets.
• Added function to compute the dual of a FormalContext.
• Now one can create a FormalContext from a CSV, CXT or RDS file directly, without needing to “load()” it.
• FormalContexts can now be saved to CXT format, in addition to RDS.
• Added functions to compute the top and the bottom of a concept lattice.
• Added new function sub() to extract a single Concept from a ConceptSet.
• Added functions %holds_in% and %respects%, which check the validity of a set of implications in a formal context, and if a list of attribute sets respect an implication set.
• Added functions %entails% and %~% to check the entailment and equivalence between two implication sets.
• Added new convenience function to map attributes between Sets, so computing intents, extents and closures is more robust.
• Added new functions %&% and %|% that compute the intersection (logical and) and the union (or operation) on Sets.
• Conceptual scaling, including nominal, ordinal, interordinal, biordinal and interval scales for many-valued formal contexts. Also, computation of background knowledge from the applied scales and of the implications that hold in the formal context. Added new vignette.

Breaking changes:

• The former SparseSet and SparseConcept classes are now named Set and Concept. Thus, to create an object of these types, just use Set$new(…) or Concept$new(…). Analogously, the former function as_SparseSet() is now as_Set().

Bugfixes:

• Minor bugfixes in several functions.

fcaR 1.0.7

Bugfixes:

• Minor bugfixes in several functions.
• Fixed C source that could generate some problems.

Enhancements:

• Reduced number of dependencies.
• Better printing of FormalContexts, specially for binary FormalContexts.

fcaR 1.0.6

Bugfixes:

• Detect if R has been built with no long double capabilities and warn the user if she tries to plot FormalContexts and ConceptLattices.

fcaR 1.0.5

Bugfixes:

• Fixes #17. The reduced set of implications after performing closure with reduce == TRUE now stores the matrix I, so it can be re-exported to the arules format.

Enhancements:

• More efficient version of the simplification rule. Now it performs linearly on the number of implications.
• Added a way to extend the equivalence rules by means of the registry package. This has been used to introduce the “Right Simplification” and the “Reorder” rules.
• New vignette to show how to extend equivalence rules in the package.
• More efficient version of the NextClosure algorithm to mine concepts and implications.

fcaR 1.0.4

Bugfixes:

• Fixes #15. A problem with version 1.6-6 of arules.
• Bugfix when using the trivial context (S, S, !=)
• Several bugfixes in LaTeX output.

Enhancements:

• Added export of plots to TiKZ to be included in LaTeX documents.
• Faster computation of concepts and implications.
• Added new function to decompose a concept in its irreducible components.
• Added function to compute the canonical basis from an ImplicationSet.

fcaR 1.0.3

• Changes to some C functions since they could potentially give a segfault.

fcaR 1.0.2

• Made minor changes in the DESCRIPTION.

fcaR 1.0.1

• Made changes suggested by CRAN:

  • Added examples in the documentation.
  • Changed cat() to message() in functions not related to printing.
  • Added proper contributors.

• Updated vignettes

fcaR 1.0.0

• Added a NEWS.md file to track changes to the package.