This repository contains code and data for my article "Learn Japanese Onomatopoeia with Neo4j". The chatbot is an adaptation from Tomaz Bratanic's Context-Aware Knowledge Graph Chatbot With GPT-4 and Neo4j. A big thank to Tomaz.
-
The scripts are for data download, formatting and the chatbot.
-
The data folder contains the CSV downloaded from the JapanDict and Onomato Project.
Neo4j Desktop or AuraDB
Gemini Cloud account
OpenAI account
-
run download_onomatopoeic.ipynb to download the vocabulary data. The data nodes_with_examples_final.csv is also included in the data folder
-
run onomatopoeia_embedding.ipynb to run the ada-002 embedding, cosine similarity and format the nodes. The output data are nodes(1).csv and edge(1).csv in the data folder
-
Import the data into Aura/Gemini Cloud
-
Set up the chatbot. Set up a conda environment (requirements.txt) and run the chatbot:
streamlit run chatbot.py
- Sixing Huang - Concept and Coding
This project is licensed under the MIT License - see the LICENSE file for details