The aim of this project is to implement a persistent memory in a conversational AI, or chatbot. This is done by incorporating external and internal memory structures that can maintain and access information over an extended period of interaction. By developing a chatbot with a persistent memory, we aim to achieve more engaging, context-aware, and human-like conversations.
The implementation of persistent memory in this project comprises two main components:
-
Dialogue History as Memory: This involves treating the past dialogue interactions as a form of memory. This method allows the chatbot to refer back to prior exchanges in the dialogue and enables the bot to maintain context throughout the conversation.
-
Knowledge Graph as Memory: This involves storing knowledge in an external structure, such as a knowledge graph, which the chatbot can query to retrieve information. This method allows the chatbot to possess and use knowledge that extends beyond the current conversation.
The project incorporates these two methods, allowing the chatbot to effectively utilize the benefits of both the context-awareness provided by the dialogue history and the vast, organized information available through the knowledge graph.
We welcome contributions to this project. Whether it be adding features, finding and fixing bugs, or simply improving documentation, all are greatly appreciated.
Before making any changes, please first discuss the change you wish to make via issue, email, or any other method with the owners of this repository.
This project is licensed under the terms of the CC BY-NC-SA license.
For more information, please refer to the LICENSE file.