ElectreCredibility - computes credibility matrix using procedure which is common to the most methods from the Electre family. The key feature of this module is its flexibility in terms of the types of elements accepted as input - concordance or discordance may be a result of the following comparisons: alternatives vs alternatives, alternatives vs boundary profiles and alternatives vs central (characteristic) profiles, and discordance may be provided both in non-aggregated ('partial') and aggregated form (think of discordances in Electre TRI and Electre Is methods). http://github.com/xor-xor/electre_diviz Alternatives to consider. Concordance matrix with aggregated concordance indices (i.e. one index per pair of alternatives or alternatives/profiles). Discordance indices provided in non-aggregated ('partial') or aggregated form. In the former case, 'with_denominator' parameter (see below) should be set to 'true'. Definitions of profiles (boundary or central) which should be used for classes (categories) representation. A set of parameters provided to tune up the module's operation. %1 %2 %3 %4 ]]> This parameter specifies the type of elements provided for comparison. Choosing 'boundary_profiles' or 'central_profiles' requires providing inputs 'classes_profiles' and 'profiles_performance_table' as well (which are optional by default). alternatives vs alternatives alternatives alternatives vs boundary profiles boundary_profiles alternatives vs central (characteristic) profiles central_profiles item0 This parameter specifies if you want to take into account only sufficiently great discordance (by using '1 - d_j(a, b) / 1 - C(a, b)' instead of '1 - d_j(a, b)' in credibility's equation). In case of selecting this option, only discordance indices for which d_j(a, b) > C(a, b) holds are taken into account. true Use only maximum discordance index in discordance aggregation part of credibility's equation (i.e. '1 - max(d_j(a, b))'). When 'with_denominator' is set to 'true', this parameter is ignored. false This parameter specifies whether the discordance indices are provided in non-aggregated form. true

Credibility matrix computed from the given data.

Messages or errors generated by this module.