TransLIME: Towards transfer explainability to explain black-box models on tabular datasets

Rehan Raza; Guanjin Wang; Hamid Laga; Kok Wai Wong; Wolfgang Nejdl

doi:10.1016/j.ins.2025.122891

Back

TransLIME: Towards transfer explainability to explain black-box models on tabular datasets

Journal article

Open access

Peer reviewed

TransLIME: Towards transfer explainability to explain black-box models on tabular datasets

Rehan Raza, Guanjin Wang, Hamid Laga, Kok Wai Wong and Wolfgang Nejdl

Information sciences, Vol.730, 122891

2026

DOI: https://doi.org/10.1016/j.ins.2025.122891

Appears in Open Access via Read & Publish Agreements

Files and links (1)

pdf

Published3.30 MBDownload View

CC BY-NC V4.0, Open Access

Abstract

Explainable AI

LIME

Model agnostic explanations

Model transfer

Transfer learning

Explainable Artificial Intelligence methods have gained significant traction for their ability to elucidate the decision-making processes of black-box models, particularly in high-stakes fields such as healthcare and finance. Among these, Local Interpretable Model-agnostic Explanations (LIME) stands out as a widely adopted post-hoc, model-agnostic approach that interprets black-box predictions by constructing an interpretable surrogate model on perturbed instances to approximate the local behavior of the original model around a given instance. However, the effectiveness of LIME can depend on the quality of the training data used by the black-box model. When trained on limited or low-quality data, the black-box model may yield inaccurate predictions for perturbed samples, resulting in poorly defined local decision boundaries and consequently unreliable explanations. This limitation is especially problematic in data-scarce settings. To overcome this challenge, we propose TransLIME, a novel end-to-end explainable transfer learning framework that improves the local fidelity and stability of LIME on limited tabular datasets by transferring relevant explainability knowledge from a related auxiliary source domain with a shifted distribution. Also, in TransLIME, only representative source prototype explanations obtained through clustering are transferred to the target domain, thereby reducing cross-domain exposure of both data and explanatory information during transfer. Experimental evaluations on real-world datasets demonstrate the effectiveness of the proposed framework in improving explanation quality in target domains with limited data.

Details

Title: TransLIME: Towards transfer explainability to explain black-box models on tabular datasets
Authors/Creators: Rehan Raza - Murdoch University
Guanjin Wang - Murdoch University, School of Information Technology
Hamid Laga - Murdoch University, School of Information Technology
Kok Wai Wong - Murdoch University, School of Information Technology
Wolfgang Nejdl - Leibniz University Hannover
Publication Details: Information sciences, Vol.730, 122891
Publisher: Elsevier Inc.
Number of pages: 21
Identifiers: 991005833799607891
Murdoch Affiliation: Centre for Biosecurity and One Health; School of Information Technology; Centre for Healthy Ageing
Language: English
Resource Type: Journal article

Metrics

4 File views/ downloads

17 Record Views

1 Times Cited - Web of Science

InCites Highlights

These are selected metrics from InCites Benchmarking & Analytics tool, related to this output

Collaboration types: Domestic collaboration; International collaboration
Citation topics: 6 Social Sciences; 6.185 Communication; 6.185.2797 AI Ethics
Web Of Science research areas: Computer Science, Information Systems
ESI research areas: Computer Science