Author

Hassan Hassan

1 paper indexed here

Fetches their full publication history.

Not the right person? Other researchers publish under this name.

Open access Jul 2026

Robust Detection of XSS and SQL Injection in JSON-API Traffic via JSON-Aware Canonicalization and Few-Shot Adversarial Fine-Tuning

Deep-learning detectors for web attacks perform strongly on clean data yet remain brittle to semantics-preserving obfuscations, especially  in JSON centric API traffic where identical malicious intent can appear under varied encodings, Unicode confusables,  whitespace/comment patterns, or JSON carrier forms. We present a deployment-minded pipeline that integrates: (i) JSON-aware  canonicalization that standardizes benign representational variance while preserving attack-relevant semantics; (ii) multi-seed training  of BiLSTM-Attention and CharCNN detectors; (iii) black-box robustness evaluation using a bounded reinforcement-learning (RL)  adversary and per-mutator Attack Evasion Rate (AER); (iv) few-shot adversarial fine-tuning (FSAT) on curated hard cases; and (v)  temperature scaling with per-class thresholds, together with cloud-style latency profiling. On a three-class corpus stratified into  Train/Validation/Test = 173,872 / 37,258 / 37,259 and augmented with JSON-native payloads, an RL set (6,821 train / 2,882 holdout), and  a few-shot set (n = 1,499; 48.3% JSON), the BiLSTM-Attention model with canonicalization achieves clean-test macro-F1 = 0.9879 ±  0.0007, precision = 0.9889 ± 0.0007, recall = 0.9871 ± 0.0007, and accuracy = 0.9875 ± 0.0007 across three seeds. For the selected robust  configuration, per-class threshold calibration (0.4 / 0.6 / 0.6) yields validation macro-F1 = 0.9881. Canonicalization reduces AER across  common mutator families, with the largest reductions on JSON wrapping and on multi-step chains such as homoglyph ? URL-encode ? whitespace. After FSAT, the calibrated canon_on model attains clean-test precision = 0.9891, recall = 0.9870, macro-F1 = 0.9880, and  accuracy = 0.9877, while reaching RL-holdout precision = 1.0000, recall = 1.0000, F1 = 1.0000, and accuracy = 1.0000 against the evaluated  bounded black-box RL attacker. Median latency at 8 KB is 79.6 ms (P95 95.8 ms), indicating practical feasibility for API-gateway  deployment. 

Ali Taqi, Hassan Hassan, Ghaidaa Abdulsaheb · 0 citations