Ontology-based Interpretable Machine Learning for Textual Data
This work addresses the need for better interpretability in machine learning for textual data, which is important for users in domains requiring transparent AI decisions, though it appears incremental as it builds on existing interpretability methods.
The paper tackles the problem of generating interpretable explanations for black-box models on textual data by using an ontology-based sampling technique and a learnable anchor algorithm to handle complex text. The result shows that their approach produces more precise and insightful explanations compared to baselines on two real-world datasets.
In this paper, we introduce a novel interpreting framework that learns an interpretable model based on an ontology-based sampling technique to explain agnostic prediction models. Different from existing approaches, our algorithm considers contextual correlation among words, described in domain knowledge ontologies, to generate semantic explanations. To narrow down the search space for explanations, which is a major problem of long and complicated text data, we design a learnable anchor algorithm, to better extract explanations locally. A set of regulations is further introduced, regarding combining learned interpretable representations with anchors to generate comprehensible semantic explanations. An extensive experiment conducted on two real-world datasets shows that our approach generates more precise and insightful explanations compared with baseline approaches.