DEJAN releases open-source sentiment model: 7-level scoring, 84% exact accuracy, laptop-friendly
dejanseo · x · 2026-11-02
Dan Petrovic of SEO firm DEJAN has released a new sentiment classification model, free on Hugging Face as a gift to the SEO community.
- Capability: chunks pages into headings, list items and sentences, scoring each across 7 sentiment levels from very positive to very negative; hits 84% of labels exactly and 95% within one point, and is small enough to run on a laptop.
- Three ways to use it: a public demo at dejan.ai/tools/sentiment/ (paste a URL or text, share results via link), integration in the DEJAN App under Pages → Sentiment with per-project saved runs, or a Hugging Face download with a quick-start script that splits and scores a page.
- Output: word-count-weighted page sentiment from +3 to −3, page stats (words, reading time, Flesch reading ease), a color-coded heatmap, and a sortable, exportable per-chunk table.
More from Models
- Tuning 0.001% of Gemma 3 270M's weights lifts sort accuracy from 1% to 97% — dejanseo · 2026-11-02
- SemiAnalysis launches GDPx benchmark: best model Claude Opus 5.5 scores just 12% — zephyr_z9 · 2026-11-02
- Asianometry Deep Dive: How Zhipu AI's GLM Became China's Enterprise AI Champion — wren6991 · 2026-11-02
- Design work head-to-head: Opus 5.5 crushes astra, 'creative mad genius vs junior designer' — tengyanAI · 2026-11-02
- Grok 4.7 tops Investment Arena live-trading contest at +3.9%, beating the S&P 500 — my_cat_can_code · 2026-11-02
- Power users hit weekly AI limits within hours, fueling subscription value debate — RachelVT42 · 2026-11-02