EU Consultation
Dashboard
About
Figures
Search
Review
Explorer
Log in
βοΈ
Tobacco Taxation Excise Duties For Manufactured Tobacco Products Updated Rules
03 September 2025 β 31 October 2025
View on EU portal β
Last model run:
2026-04-02T00:40:42
β
Total submissions
18478
03 Sep 2025 β 31 Oct 2025
β
Against proposal
88.6%
16364 entries
β
Countries represented
88
887 duplicates (4.8%)
β
Manual labels collected
4850
by 4 reviewer(s)
β
Submissions with attachment
1.5%
279 with attachment · 18199 text-only
β
Stance model
18 runs
accuracy 98.7% · F1 98.7%
latest: 88.6% against · 11.4% for · 0.0% unclear
1988 pseudo-labels · 2026-04-02 00:32
avg confidence ·
against 99.9%
·
for 99.5%
·
unclear 52.6%
manual agreement: 93.7% (4669 reviewed)
β
Stance distribution
88.6%
Against
(16364)
11.4%
For
(2109)
0.03%
Unclear
(5)
β
Model training progress
88.56%
Against
11.41%
For
0.03%
Unclear
Β·
98.69%
Eval acc.
94.29%
Manual agree
β
Submission volume
β
Submitter types
Stance breakdown
β
Stance by submitter type
β
Stance over time
Stance by country (top 10)
βΉ
βΊ
β
β
β
Who supports the proposal? (% For by type)
β
Who opposes the proposal? (% Against by type)
Geographic distribution
Top 10 countries
βΉ
βΊ
β
β
β
Submissions by country
β
Submissions by country over time
Content analysis
Top keywords
βΉ
βΊ
β
β
Top phrases (bigrams)
βΉ
βΊ
β
β
Submissions by language
βΉ
βΊ
β
β
β
Submission length
median 152 words
β
Stance by submission length
Duplicate analysis
β
Near-duplicate rate by submitter type (β₯95%)
β
Text duplicate group sizes
largest: 12
β
Near-duplicate count by submitter type (β₯95%)
Most-submitted duplicate attachments
(max: 6)
βΉ
βΊ
β
β
Bot & coordinated submission detection
β
High semantic similarity
(β₯95%)
3.3%
602 of 18478 entries
β
Time burst entries
(β€300 s)
95.1%
17572 entries in 652 burst groups
β
Largest burst group
3922
avg 1162.3 entries per burst
β
Submissions & high-similarity entries over time
(β₯95%)
β
Semantic similarity distribution
Similarity patterns
β
Burst size vs similarity
β
Similarity over time
(daily median Β± IQR)
β
Similarity distribution by stance
(% within stance)
Submission timing
β
Submissions by day & hour β all
(CET/CEST β because the EU still hasn't fixed daylight saving)
β
Submissions by day & hour β EU-27 only
(CET/CEST β the EU should just pick UTC+1.5 and disappoint everyone equally)
β
Hourly pattern β private vs corporate, all
β
Hourly pattern β private vs corporate, EU-27