
HanziFinder
HanziFinder is a powerful substructure search tool specifically designed to search Chinese characters. In the future, HanziFinder's technique can be expanded to find all substructures within a character database which has been recreated using our proprietary indexing method. HanziFinder leverages a comprehensive database of nearly 89,000 characters to efficiently identify all characters containing specific substructures—such as "乃" within "奶" and "孕".
Our patent-pending, graph-based search algorithm systematically analyzes unlabeled characters from a continuous data stream, categorizing them into positive or negative examples based on substructure presence, rapidly enriching the dataset. This enables linguists, researchers, and language enthusiasts to quickly explore structure-meaning relationships among Chinese characters with precision and ease.

Jennifer Ball
ORIGINAL IDEA
