The honesty metric
of public opinion
Likes and views show only loudness. PJQ measures the direction of the shock. We analyze YouTube comments to show how an audience truly received a video — separating substantive debate from fan-noise and bot farms.
What PJQ actually measures
PJQ reads the comments under a video and assigns each one to exactly one of seven stances: substantive support, disagreement, neutral remark, off-topic, fan praise, inorganic activity and 'own agenda'. Mood and support scores are computed only from substantive opinions — praise without an argument, spam and self-promotion are kept as separate counters and never inflate the score. The result is a read on how the audience received the material on the merits, not on how loud the noise around it was.
| Question to the data | How PJQ answers |
|---|---|
| Is 'I agree' separated from 'I love the author'? | Yes — substantive support (SUP) and praise without an argument (THIN) are counted separately |
| Are bots and inorganic activity visible? | Yes — inorganic activity gets its own SUS cluster and stays out of the opinion scores |
| What about comments pushing their own agenda? | Crypto-promo, self-promotion and link-dropping are split into AGN — a separate counter |
| Does noise affect the support score? | No — THIN/OFF/SUS/AGN are not part of the Mood-Net formula |
| Can topics be compared across the media field? | Yes — 11 fixed domains, one rubric across all topics |
| Is raw user data stored? | No — de-identified aggregates only, no user dossiers |
Seven layers of audience resonance
We strictly assign each comment to exactly ONE of the seven categories. The consensus index (Mood-Net) is recalculated only from substantive opinions.
AFFECT THE PURE CONSENSUS INDEX (Opinion Pool)
These categories form the core of the discussion. Only here viewers engage with the author's claims on the merits.
Support
Agreement with the author's thesis, scientific or argued defense of the position.
Against
Constructive disagreement, fact-checking, debunking, counter-arguments.
Neutral
Constructive on-topic questions, cautious balanced takes without strong emotion.
EXCLUDED FROM THE FORMULA (separate counters)
Noise, fan adoration, spam raids. These comments boost engagement but dilute quality.
Thin Positive
Emotional cliches ('amazing, super'), emoji likes, respect for the blogger without engaging with the substance.
Offtopic
Technical questions about audio, timestamps, clothes or clips, unrelated stories.
Inorganic
Ad spam, recovery scammers, bot farms, click-farming and engagement-bait.
Own Agenda
Commenters siphoning traffic to their own product, Telegram channel or crypto platform.
Top featured slices
Auto-curated by the open formula (see docs/FEATURED.md): log10(n) x signal-purity x freshness x informativeness, then diversified by domain. Click any to inspect in the Seismograph.
PJQ does not keep your personal dossiers. We don't track people. Our only job is to extract a distilled verdict of reception from the chaos of the media stream.
What is PJQ?
PJQ (Public Judgment Quotient) is a tool to analyze YouTube comments and measure real audience sentiment. Paste a video link and PJQ classifies every comment into 7 stances, scores net mood and controversy, and filters out bots, hype and spam — so you see how an audience truly received a video, not just likes and views.
How do I analyze a YouTube video's comments?
Paste the YouTube video URL into PJQ and start the analysis. PJQ samples the comments, classifies each one, and returns a verdict with the audience's net mood, a controversy score and example quotes.
Can PJQ detect bots, spam and fake hype?
Yes. PJQ separates substantive opinion from parasitic noise — bot farms, copy-paste spam and fan-hype are classified apart and excluded from the mood and support scores.
Is it free to try?
Yes — new accounts get a free first analysis. After that you top up with crypto and pay per analysis; the public Audience Trends dashboard is free to browse.