Display Device Chat Compression for Lower Cloud Server Costs
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing chatbot services on display devices incur high server usage costs due to the number of user interactions, leading to inefficient operation when few users are active, and there is a need for a method to reduce these costs while maintaining functionality.
Innovation Solution
A display device with on-device AI capabilities compresses user chat messages into abbreviated forms using language models and context information, reducing the data transmitted to cloud-based servers, thereby minimizing server usage costs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If chat messages are transmitted in full form to cloud servers, then chatbot response accuracy is improved, but server usage costs increase
Solution Approach 1:
The chat message is segmented into two parts: context information (which is compressed and sent to the server) and the actual user input (which is sent in full). This segmentation allows the server to process only the essential context while maintaining response accuracy through the complete user input.
Solution Approach 2:
The system performs preliminary compression of context information before transmitting to the server. By pre-compressing the context using language models and removing redundant information, the system reduces the data load on the server while maintaining the essential meaning needed for accurate responses.
2Loss of energy
If message compression is applied to reduce server costs, then server usage costs decrease, but information loss may occur
Solution Approach 1:
A language model acts as an intermediary to compress context information while preserving its essential meaning. The language model identifies and removes redundant information, keeping only the critical context needed for accurate chatbot responses, thus preventing information loss during compression.
Solution Approach 2:
The system creates a compressed copy of the context information that retains the essential meaning while reducing data size. This compressed copy is then transmitted to the server, ensuring that the core information is preserved without transmitting the full original message.
3Reliability
If full chat messages are transmitted, then complete information is provided to server, but data transmission volume increases
Solution Approach 1:
The system extracts and separates context information from the full chat message. Only the extracted context information is compressed and transmitted to the server, while the complete user input is sent separately. This extraction reduces the overall data transmission volume while maintaining information completeness.
Solution Approach 2:
The system changes the parameter of data transmission by compressing the context information using language models. This parameter change reduces the transmission volume of context data while preserving the essential information needed for accurate chatbot responses.
Data Source
AI summary
A display device to display content, identify a chat message input by a user while the content is displayed, change the identified chat message into an abbreviated message having a meaning identical or similar to the identified chat message, based on context information associated with the content being displayed, and transmit the abbreviated message to a server enabled to provide a chatbot response.