abc_model: Apply the ABC model for literature-based discovery with improved filtering

Description

This function implements the ABC model for literature-based discovery with enhanced term filtering and validation.

Usage

abc_model(
  co_matrix,
  a_term,
  c_term = NULL,
  min_score = 0.1,
  n_results = 100,
  scoring_method = c("multiplication", "average", "combined", "jaccard"),
  b_term_types = NULL,
  c_term_types = NULL,
  exclude_general_terms = TRUE,
  filter_similar_terms = TRUE,
  similarity_threshold = 0.8,
  enforce_strict_typing = TRUE,
  validation_method = "pattern"
)

Value

A data frame with ranked discovery results.

Arguments

co_matrix: A co-occurrence matrix produced by create_comat().
a_term: Character string, the source term (A).
c_term: Character string, the target term (C). If NULL, all potential C terms will be evaluated.
min_score: Minimum score threshold for results.
n_results: Maximum number of results to return.
scoring_method: Method to use for scoring.
b_term_types: Character vector of entity types allowed for B terms.
c_term_types: Character vector of entity types allowed for C terms.
exclude_general_terms: Logical. If TRUE, excludes common general terms.
filter_similar_terms: Logical. If TRUE, filters out B-terms that are too similar to A-term.
similarity_threshold: Numeric. Maximum allowed string similarity between A and B terms.
enforce_strict_typing: Logical. If TRUE, enforces stricter entity type validation.
validation_method: Character. Method to use for entity validation: "pattern", "nlp", "api", or "comprehensive".