1 Structured Data Tags
◦ Embeds AI readable data (e.g., facts, datasets) directly in HTML, simplifying extraction for LLMs and knowledge graphs.
2 Implicit Learning
◦ Provides AI with relationship hints (e.g., causation, correlation) via targets, confidence, direction, enhancing reasoning and graph construction.
3 Lists and Metadata
◦ Supports complex data like sequences and annotations (e.g. lists, metadata, hierarchical/nested data blocks ), enabling richer, contextual data representation.
4 Nested Blocks
◦ Allows hierarchical data structures (e.g., datasets containing facts), increasing flexibility for diverse use cases.
5 Data Cleaning Attributes (timestamp, expires, hash)
◦ Automates removal of outdated or duplicate data, improving AI training data quality.
6 Extensibility (xs:any, xs:anyAttribute)
◦ Permits custom elements/attributes (e.g., custom-attr=”value”), adapting to user specific needs without schema changes.
7 Robust Validation (aodm-validator.js)
◦ Ensures correct markup with checks for type, hash, confidence, reducing errors in AI pipelines.
8 Versioning and Priority (version, priority)
◦ Tracks schema updates and prioritizes critical data for AI processing, ensuring future-proofing and efficiency.
9 Format Specification (format)
◦ Supports varied data types (e.g., text, json, image), broadening applicability for multimedia or structured data.
10 CMS Integration (WordPress, Drupal, Joomla, Shopify, Salesforce, and HubSpot)
◦ Simplifies adoption via shortcodes/blocks (e.g., [aodm type=”fact”]), making it accessible for non-technical users.
11 Lightweight Design
◦ Minimal overhead (~100-500 bytes per tag, ~16 KB core package), maintaining fast page loads and easy distribution via nesaraq.com.
12 HTML5 Compatibility
◦ Integrates seamlessly with web pages using xmlns:aodm, requiring no additional runtime dependencies for static use.