Kratos-AI/Handwritten-Chemistry-Notes-Dataset

English Handwritten Chemistry Notes Dataset

This dataset contains high-resolution images of handwritten chemistry notes written in English. The collection includes equations, reaction mechanisms, periodic table references, structural diagrams, and descriptive explanations. It supports AI research in handwriting recognition, chemical structure understanding, and document analysis for STEM and educational applications.

Contact

For queries or collaborations related to this dataset, contact:

• [email protected]
• [email protected]

Supported Tasks

• Task Categories:

  • Image Classification
  • Text and Formula Recognition (OCR)
  • Diagram Detection
  • Document Layout Understanding

• Supported Tasks:

  • Recognition of handwritten English chemistry text and symbols
  • OCR for chemical equations and reaction mechanisms
  • Detection and classification of molecular and structural diagrams
  • Analysis of handwritten content combining text, symbols, and arrows
  • Research in AI-driven chemistry education and note digitization

Languages

• Primary Language: English
• Secondary Presence: Chemical symbols and notations (e.g., H₂O, NaCl, CH₃COOH, →, Δ, etc.)

Dataset Creation

Curation Rationale

The dataset was created to improve AI systems that interpret handwritten chemistry notes combining symbolic, textual, and graphical information. It is designed for research in chemistry education, automated content transcription, and intelligent study-assist tools.

Source Data

• Contributors: Students, educators, and volunteers providing personal or academic chemistry notes
• Collection Process: Notes were scanned or photographed. All identifiable personal data were removed prior to inclusion.

Other Known Limitations

• Bias: Mostly academic and classroom-level notes
• Variability: Differences in drawing styles for structures and chemical diagrams
• Scope: Focused on general, organic, and inorganic chemistry; limited advanced or research-level topics

Intended Uses

✅ Direct Use

• Training OCR and handwriting recognition systems for chemistry content
• Research in AI-based document understanding and scientific content extraction
• Educational technology applications such as smart note recognition
• Digitization of classroom and study materials for STEM learning

❌ Out-of-Scope Use

• Attempting to identify individuals through handwriting
• Reuse of individual handwriting for commercial purposes without consent
• Surveillance or behavior analysis from writing styles

License

CC BY 4.0