IDEAINNOVIDEAINNOV
Sign inSign up free
← Home

📖 User Manual

Three programs + desktop app — description and step-by-step usage for each

🧬 V134 Paper Pipeline

Description

  • Automatically derives coupled PDE systems from a natural-language query or an uploaded PDF/DOCX/HWPX.
  • Generates an academic paper in the SCI Q1 11-section structure (Intro, Problem, Hypotheses, Method, Theory, Experiments, Results, Discussion, Limitations, Conclusion).
  • Includes 47+ editable OMML equations, 8 figures, and automatic reference citations.
  • Automatically evolves unknown correction terms on a baseline equation to raise R² accuracy (the internal algorithm is patent-pending and not disclosed).
  • Uses an accelerated convergence technique for faster computation and 5-seed Bootstrap statistics for reproducibility.

How to use

  1. Input: type a research topic, or upload a source paper (PDF/DOCX/HWPX, ≤20MB / 30 pages).
  2. Pick a domain (reaction-diffusion, materials, battery, fluids...) to improve library matching.
  3. Click "Generate" — governing-equation derivation → evolution → simulation → paper writing run automatically.
  4. Review: a virtual SCI Q1 reviewer iterates up to 3 rounds, showing scores and fixes.
  5. Download the finished DOCX (OMML equations included). Output is kept in your archive for 3 days.

💼 buz_engin Marketing Engine

Description

  • Auto-generates PR/marketing proposals and campaign content from an RFP or a marketing topic.
  • Extracts 10 core keywords to design target, channel, and message strategy.
  • Provides per-channel content examples (copy, hashtags) for Instagram, YouTube, blogs, etc.
  • Derives 3 Win Themes and KPIs (with rationale) to strengthen the winning strategy.
  • Produces a 40–80 page PPTX proposal in the Impact-8 structure.

How to use

  1. Input a marketing topic or upload an RFP document.
  2. Keyword analysis: 10 core keywords and target insights are extracted automatically.
  3. Choose a direction among the proposed Win Themes / channel strategies.
  4. Generate campaign content examples and a PPTX proposal.
  5. Download the PPTX for editing and presentation.

📑 HWPX ↔ DOCX ↔ PDF Converter

Description

  • Converts between Hangul (HWPX), MS Word (DOCX), and PDF while preserving style, equations, and tables.
  • Equations become editable OMML objects; tables keep their layout.
  • Extracts figures/diagrams from PDFs and inserts them into the output (figure parsing).
  • Measures visual fidelity (SSIM) to verify conversion quality against the original.
  • File limits: ≤20MB / 30 pages (Pro 40MB/100p, Enterprise 100MB/200p).

How to use

  1. Upload the file to convert (HWPX/DOCX/PDF).
  2. Select the target format (e.g., PDF→DOCX).
  3. Run conversion — style, equations, and tables are preserved.
  4. Check the SSIM score and preview against the original.
  5. Download the result. Oversized files are split or handled via a higher plan.

🖥️ IDEAINNOV Desktop (Windows app)

Description

  • Your query history is stored locally on your own PC (local SQLite), not on the server.
  • A personal output vault auto-collects and organizes your papers/conversions and verifies sha256 integrity.
  • Your login session persists, so you don’t sign in every time.
  • Hardened with renderer isolation, DPAPI-encrypted token, and blocked execution of network-origin files.
  • Runs on Windows 10/11 (64-bit).

How to use

  1. Download (currently beta): email info@ideainnov.com to request the installer. A public download will open once code-signing is complete.
  2. Install, launch the app, and sign in with your ideainnov.com account as usual (the session is saved).
  3. Use: run papers/proposals/conversions just like on the web; results are auto-saved to your local vault, and you can reopen past work from the History window (menu → History).
  4. Upgrade: at login the app shows a notice if a newer version exists (clicking it opens the official download page in your browser). You can also check manually from the “Check for updates” menu. For security the app never downloads or runs a file by itself.

Beta request · contact: info@ideainnov.com /desktop →

🧩 Hard mode & iterative interpretation

Description

  • A "hard problem" (난제) is a genuinely difficult engineering/math/science topic — one that resists standard methods or yields low accuracy.
  • Hard mode derives deeper governing equations with coupled models, symbolic regression, and multi-stage validation to find an improved solution.
  • It is heavier to compute and validate, so it costs 4 credits (vs 1 for a regular paper).

Improvement per round

  • Iterative interpretation re-solves the same problem n times, each round starting from the previous solution to refine the governing equation further.
  • Accuracy (R²) rises with each round but gradually converges — it does not improve forever; the gains shrink toward a limit.
  • Credits scale with rounds: hard × n = 4×n (e.g. 3 rounds = 12, 5 rounds = 20).
0.9800.9850.9900.995baseline 0.9820.99512345Interpretation roundAccuracy R²
Per-round improvement (R²) — illustrative. Starting from a single-physics baseline, iterated coupled interpretation improves quickly at first, then converges.