Using NLP phrase extraction on Chinese-language censored pages, the system discovered 1,125 new censored domains not present on any publicly available blocklist, producing a list 12.5× larger than the standard Citizen Lab list (220 web pages, 85 domains). Across three evaluations (unigrams, bigrams, trigrams, each capped at 1,000,000 URLs), only 3 of the top 50 discovered domains overlapped with FilteredWeb's top 50.

China actively censors websites far outside the popular-traffic tier: many discovered censored domains appear in the tail of the Alexa Top 1,000,000, and some are absent from Alexa entirely. This demonstrates the GFW pursues content-classified hosts regardless of traffic rank, not only high-visibility platforms.

§5.2, Figure 3 detection

The 1,125 newly discovered censored domains span a broad taxonomy: Chinese human rights organizations, Tibetan rights outlets, Falun Gong and religious freedom sites, minority news, privacy-enhancing technology providers, and sources covering Tiananmen and the 1989 democracy movement—none appearing on the Alexa Top 1,000 or FilteredWeb's blocklist. Privacy-enhancing technology providers appear explicitly as a censored category alongside political and religious content.

§1, §5.1, §6 detection

Multi-word Chinese phrases as search seeds discover qualitatively different censored sites than individual English words: the phrase 'Chinese human rights violation' surfaces Chinese activist homepages and culture-specific outlets, while individual constituent words return only well-known Western media. TF-IDF scoring against a Chinese corpus ranks culturally rare phrases (e.g., '自由亚洲电台' / Radio Free Asia) as high-signal seeds and discards common filler phrases.

§3.1–3.2 evaluation

Culturally specific Chinese phrases are strong predictors of censorship: unigrams for controversial figures—Wang Qishan (74%), Li Hongzhi (64%), Guo Boxiong (62%), Hu Jintao (56%)—returned the highest block rates. Trigrams such as 'Beidaihe meeting' (54%), 'CCP's religious policy' (42%), and 'Tiananmen Square demonstrations' (32%) showed similar patterns, confirming King et al.'s finding that references to collective political dissent are disproportionately targeted.

§5.3, Tables 2–4 detection

Of 6,915,266 domains extracted from the 572 GiB Geedge Networks Leak (GNL), 298,955 censored domains (93.7% of all GNL-censored domains) appear in neither Tranco top-1M nor CitizenLab test lists. Measurements across China (Guangzhou/Nanjing), Myanmar, Pakistan, and Algeria confirmed censorship via DNS injection and SNI-based TLS connection termination. The GNL covers 25–62% of Tranco-censored domains across countries, showing substantial but incomplete overlap. This vendor-side ground truth reveals a censorship surface roughly two orders of magnitude larger than curated academic test lists.

2026-sheffey-geedge §4.1, Table 2 dpisni-blockingdns-poisoningmeasurement-platform

Majority-vote ML inference (OCSVM + IF) over OONI data uncovered at least 5 previously undocumented DNS injection IPs active in Russia (e.g., 195.19.90.226, 95.167.13.51, 61.95.167.13.50, 188.19.132.154, 144.85.142.29.248) absent from OONI's existing blocking-fingerprints database, along with novel fingerprints in Italy, Czech Republic, and the UK. Records with fewer than 50 instances were excluded as a conservative false-positive filter.

2024-calle-toward §4.3, Table 5 dns-poisoningml-classifiermeasurement-platform

XGBoost trained on a single month of OONI data achieves near-optimal performance; expanding the training window to 24 months produces deviations of only 0–5 percentage points for FNR, 0.07 PP for FPR, and 0.10 PP for accuracy — suggesting that larger windows introduce noise and overfitting rather than improving detection. Isolation Forest performance degrades more sharply, with accuracy dropping ~5 PP as training data grows beyond 6 months.

2024-calle-toward §4.2, Figure 3 dns-poisoningml-classifiermeasurement-platform

For the Isolation Forest model, resolver ASN (SHAP importance 0.237) and probe ASN (0.220) are the two most predictive features for DNS tampering, reflecting that censorship is topologically concentrated at specific network vantage points. For XGBoost, headers_match dominates (0.317), followed by asn_control_match (0.177), indicating that supervised models rely more on cross-layer consistency signals. DNS tampering represents only 0.5–0.8% of all OONI measurements across 2022–2023 (Figure 2), creating severe class imbalance in any training set.

2024-calle-toward §4.1, Table 4, Figure 2 dns-poisoningml-classifiermeasurement-platform

XGBoost achieves a False Positive Rate of 0.0005, True Positive Rate of 0.9403, and overall accuracy of 0.9991 on OONI global DNS measurement data (2.5% stratified sample), vastly outperforming unsupervised alternatives: Isolation Forest achieves FPR 0.1321 / ACC 0.8699, and One-Class SVM degrades to FPR 0.9711 / ACC 0.0598, making OCSVM effectively unusable for this task.

2024-calle-toward §4.1, Table 3 dns-poisoningml-classifiermeasurement-platform

Brown et al. (2023) combined supervised ML models trained on expert-labeled data with unsupervised models establishing a baseline of 'normal' behavior to detect DNS-based censorship from Satellite and OONI datasets, achieving high true-positive rates for both known and new DNS censorship instances. The hybrid supervised/unsupervised approach is proposed as a template for the LLM-based system.

2024-gao-extended §2 Related Works dns-poisoningmeasurement-platformml-classifier