About Us


Original Idea

Jennifer

Jennifer Ball is a published author (Higher Math the Book Moose Minnion Never Wrote, Faber, 1991; Catalyst, Faber, 1997; they were also translated and published in China in Chinese in 2019 and 2017, respectively). She also co-wrote The Verbum Book of Digital Typography (1990), and has two published linguistic papers on Chinese Hanzi: "Meow" is just another name for "Cat" (2019) and The Role of the “Reproduction” 乃 Character in Chinese Writing: 282 Characters and 59 Definitions (2023)

Jennifer is nearly fluent in Hanyu and Hanzi (Mandarin and Chinese writing, respectively) after studying them for 16 years. She has written 300 pages of a new novel focused on teaching people written Chinese. Bias creates a pattern that we could be harnessing to learn languages faster.

Xu Chao, a former student, and Jennifer together made the first substructure search engine of Chinese characters: HanziFinder.com.

Implementation

Xu Chao

Xu Chao holds a bachelor’s degree in Nanomaterials and Technology from Soochow University and later pursued research in semiconductor device physics at FUNSOM. His work was published in Applied Physics Letters, including: Organic Thin-Film Memcapacitors and Carrier Injection in Organic Electronics: Injection Barrier Control at Metal/Organic Interfaces

Inspired by molecular representation and substructure matching methods in computational chemistry, he innovated on Hanzi structural modeling and substructure search algorithms and engineered them into practical tools.

He later focused on drug discovery and AI engineering, spanning molecular property prediction, molecular generation, and AI platform development.

Patent

HanziFinder's patent application was approved March 24, 2026: SYSTEMS AND METHODS FOR REPRESENTING AND SEARCHING CHARACTERS. Patent No.: US 12,586,401 B2

Abstract: Methods and supporting systems for representing and searching characters, comprising: obtaining an image of a character, labelling a structure of the character by defining a plurality of nodes and a plurality of edges on the character in the image, and generating a representation of the character by extracting a set of two-dimensional coordinates to represent the plurality of nodes and by extracting a matrix to represent the plurality of edges, and providing the representation in a searchable database.

Preview