Abstract :Chinese patterns often rely on "Rebuses"—visual puns where images (e.g., "Bat") symbolize concepts (e.g., "Fortune") via homophony. Current metadata fails to capture this phonetic link. We present a prototype visualization tool powered by a 27,000-item dataset, using a "Phonetic Bridge" graph to reveal these hidden cultural meanings to global audiences.

Keywords:Visual Rebus; Homophony; Knowledge Graph; Cultural Visualization; User Interface Design

1. Introduction: The Barrier of the "Silent Text"

The theme of DH2026 is "Engagement." However, meaningful engagement with Traditional Chinese Patterns (TCP) is often blocked by a unique linguistic mechanism: the Rebus (visual pun). In Chinese visual culture, images are frequently selected not for their physical attributes, but for their phonetic values.

A classic example is the Bat (Bian Fu). To a Western viewer, a bat might symbolize darkness or horror. In Chinese tradition, because Fu (Bat 蝠) sounds exactly like Fu (Fortune 福), the bat is the primary icon of blessing. Similarly, a Cat (Mao) and a Butterfly (Die) depicted together are not a biological scene, but a prayer for longevity, as Mao-Die sounds like "Octogenarian" (Mao Die 耄耋).

Current digital collections and museum databases fail to engage users with this layer of meaning. A keyword search for "Fortune" rarely returns images of "Bats" because the metadata describes the Visual Object (What is seen), not the Phonetic Bridge (What is heard), nor the Implied Meaning (What is understood). This short paper presents a work-in-progress tool designed to bridge this gap: the "Rebus-Graph Interface."

2. Data Foundation: Mining the Puns

This tool is not built on empty theory but is powered by a massive, manually annotated dataset of 27,000 Traditional Chinese Patterns.

While the full dataset covers morphology and history, for this specific tool, we extracted a specialized sub-graph focused on the "Auspicious Meanings" (Ji Xiang Yu Yi) category.

We structured the data into tri-nodes:

Visual Signifier (S): The object depicted (e.g., Vase, Quail, Lotus).

Phonetic Value (P): The pronunciation in Mandarin (e.g., Ping, An, Lian).

Semantic Signified (I): The abstract concept (e.g., Peace, Safety, Continuity).

3. Methodology: Constructing the "Phonetic Bridge"

The core innovation of our tool is the explicit modeling of the "Homophonic Link" edge type in a Knowledge Graph.

Most semantic networks connect nodes based on ontological hierarchy (e.g., Bat --is a--> Mammal).

Our network connects nodes based on phonetic proximity (e.g., Bat --sounds like--> Fortune).

We developed a prototype interface using Neo4j and D3.js that allows users to traverse these hidden paths. The system logic operates as follows:

Input: User selects an image of a "Halberd" (Ji) and a "Chime Stone" (Qing).

Process: The system retrieves the phonetic values.

Halberd →Ji

Chime → Qing

Bridge: The system queries the homophone dictionary.

Ji ≈ Ji (Auspicious/Luck)

Qing ≈ Qing (Celebration)

Output: The system displays the hidden meaning: "Ji Qing You Yu" (Abundance of Luck and Celebration).

4. The Interface: Visualizing the Invisible

The Short Paper presentation will feature a live (or recorded) demo of the interface, highlighting two key visualization modes:

Mode A: The "Decoder Lens"

Designed for museum contexts. A user uploads an image of a complex pattern (e.g., "Three Goats"). The tool identifies the object ("Goat" / Yang) and overlays a "Phonetic Bubble" showing the link to "Sun" (Yang), revealing the idiom San Yang Kai Tai (Three Suns/Goats Bring Prosperity). This visualizes the cognitive leap that a native speaker makes instantly, making it accessible to non-native speakers.

Mode B: The "Pun Explorer"

A networked view where users can explore the ambiguity of symbols. Clicking on the node "Lian" explodes into a cluster showing that it links visually to "Lotus" (Lian) but semantically to both "Continuous" (Lian) and "Incorruptible" (Lian). This effectively maps the "poetic space" of the Chinese language through visual icons.

5. Case Study: The "Vase" (Ping)

We will demonstrate the tool's efficacy using the "Vase" cluster from our dataset.

Scenario: A user searches for "Peace."

Traditional DB: Returns zero results (because the images are of Vases, Apples, and Saddles).

Rebus-Graph: The tool understands that "Peace" (Ping An) is phonetically linked to:

Vase (Ping)

Apple (Ping Guo)

Quail (An Chun)

Saddle (An)

Result: The tool retrieves patterns containing Vases and Quails, correctly identifying them as "Peace" motifs. This fundamentally changes the retrieval paradigm from "Visual Matching" to "Cultural Matching."

6. Discussion and Future Work

This tool is a prototype for "Cultural Translation." It argues that "Engagement" with world memory requires tools that can translate not just language, but logic.

Our next steps involve:

Dialect Mapping: Currently, the tool uses Mandarin. However, many Southern patterns rely on Cantonese or Hokkien puns. We plan to add a "Dialect Toggle" layer.

Generative Education: Using the graph to help users design their own auspicious patterns (e.g., "I want to wish my friend success in exams" $\rightarrow$ System suggests "Champion Vase").

By visualizing the invisible phonetic ties, we allow global audiences to "read" Chinese patterns as text, transforming a static decorative experience into an engaging linguistic puzzle.

1.Bartholomew, Terese Tse (2006): Hidden Meanings in Chinese Art. San Francisco: Asian Art Museum.

2.Eberhard, Wolfram (1986): A Dictionary of Chinese Symbols: Hidden Symbols in Chinese Life and Thought. London: Routledge & Kegan Paul.

3.Feng, Jicai (冯骥才) (2007): The Iconography of Chinese Auspicious Patterns (中国吉祥图说). Beijing: China Social Sciences Press.

4.Hyvönen, Eero (2012): “Publishing and Using Cultural Heritage Linked Data on the Semantic Web”, in: Morgan & Claypool Publishers (eds.): Synthesis Lectures on the Semantic Web: Theory and Technology. Williston, VT: 1–159.

5.Nozomi, Imano / Mitsuru, Maeda (2019): “Visualizing Hidden Phonetic Links in East Asian Art: A Computational Approach”, in: International Journal of Digital Humanities 1, 2: 187–205.

6.Rawson, Jessica (1984): Chinese Ornament: The Lotus and the Dragon. London: British Museum Publications.

7.Welch, Patricia Bjaaland (2008): Chinese Art: A Guide to Motifs and Symbols. North Clarendon: Tuttle Publishing.