Symbolic AI for XAI: Evaluating LFIT Inductive Programming for Fair and Explainable Automatic Recruitment

Machine learning methods are growing in relevance for biometrics and personal\ninformation processing in domains such as forensics, e-health, recruitment, and\ne-learning. In these domains, white-box (human-readable) explanations of\nsystems built on machine learning methods can become crucial. Inductive Logic\nProgramming (ILP) is a subfield of symbolic AI aimed to automatically learn\ndeclarative theories about the process of data. Learning from Interpretation\nTransition (LFIT) is an ILP technique that can learn a propositional logic\ntheory equivalent to a given black-box system (under certain conditions). The\npresent work takes a first step to a general methodology to incorporate\naccurate declarative explanations to classic machine learning by checking the\nviability of LFIT in a specific AI application scenario: fair recruitment based\non an automatic tool generated with machine learning methods for ranking\nCurricula Vitae that incorporates soft biometric information (gender and\nethnicity). We show the expressiveness of LFIT for this specific problem and\npropose a scheme that can be applicable to other domains.\n