# X-MAP: eXplainable Misclassification Analysis and Profiling for Spam and Phishing Detection

Source: [X-MAP: eXplainable Misclassification Analysis and Profiling for Spam and Phishing Detection](<https://www.sota2.com/research/paper/6995477609c33900fcd6d502>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-02-19T05:00:04.488Z

Authors: Qi Zhang, Dian Chen, Lance M. Kaplan, Audun J{\\o}sang, Dong Hyun Jeong, Feng Chen, Jin-Hee Cho

Published: 2026-02-17T01:46:08.000Z

## Abstract

Misclassifications in spam and phishing detection are very harmful, as false negatives expose users to attacks while false positives degrade trust. Existing uncertainty-based detectors can flag potential errors, but possibly be deceived and offer limited interpretability. This paper presents X-MAP, an eXplainable Misclassification Analysis and Profilling framework that reveals topic-level semantic patterns behind model failures. X-MAP combines SHAP-based feature attributions with non-negative matrix factorization to build interpretable topic profiles for reliably classified spam/phishing and legitimate messages, and measures each message's deviation from these profiles using Jensen-Shannon divergence. Experiments on SMS and phishing datasets show that misclassified messages exhibit at least two times larger divergence than correctly classified ones. As a detector, X-MAP achieves up to 0.98 AUROC and lowers the false-rejection rate at 95% TRR to 0.089 on positive predictions. When used as a repair layer on base detectors, it recovers up to 97% of falsely rejected correct predictions with moderate leakage. These results demonstrate X-MAP's effectiveness and interpretability for improving spam and phishing detection.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Misclassification Detection on SMS Spam class](<https://www.sota2.com/research/sota/misclassification-detection-on-sms-spam-class.md>): Misclassification Detection; SMS Spam class; AUROC: 0.9763.

- [Misclassification Detection on SMS Non-Spam class](<https://www.sota2.com/research/sota/misclassification-detection-on-sms-non-spam-class.md>): Misclassification Detection; SMS Non-Spam class; AUROC: 0.8807.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2602.15298>)

[PDF](<https://arxiv.org/pdf/2602.15298.pdf>)