Symbolic Knowledge Extraction and Injection with Sub-symbolic Predictors: A Systematic Literature Review

TL;DR

This systematic literature review analyzes 249 methods for symbolic knowledge extraction and injection in sub-symbolic predictors, aiming to improve interpretability and transparency in AI models from an explainable AI perspective.

Contribution

The paper provides comprehensive taxonomies and classifications of existing SKE and SKI methods, offering a systematic overview and identifying gaps in the current research landscape.

Findings

Classified 132 SKE methods and 117 SKI methods systematically.

Identified the presence or absence of software implementations for each method.

Provided guidance for selecting appropriate SKE/SKI techniques for practical applications.

Abstract

In this paper we focus on the opacity issue of sub-symbolic machine learning predictors by promoting two complementary activities, namely, symbolic knowledge extraction (SKE) and injection (SKI) from and into sub-symbolic predictors. We consider as symbolic any language being intelligible and interpretable for both humans and computers. Accordingly, we propose general meta-models for both SKE and SKI, along with two taxonomies for the classification of SKE and SKI methods. By adopting an explainable artificial intelligence (XAI) perspective, we highlight how such methods can be exploited to mitigate the aforementioned opacity issue. Our taxonomies are attained by surveying and classifying existing methods from the literature, following a systematic approach, and by generalising the results of previous surveys targeting specific sub-topics of either SKE or SKI alone. More precisely, we analyse 132 methods for SKE and 117 methods for SKI, and we categorise them according to their purpose, operation, expected input/output data and predictor types. For each method, we also indicate the presence/lack of runnable software implementations. Our work may be of interest for data scientists aiming at selecting the most adequate SKE/SKI method for their needs, and also work as suggestions for researchers interested in filling the gaps of the current state of the art, as well as for developers willing to implement SKE/SKI-based technologies.