Client Device AI Model for Group Message Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Users face challenges in managing and storing group multimedia messages on mobile devices due to the sheer volume of data, with many messages being irrelevant, leading to memory issues and the need for manual deletion, while also wanting to avoid cloud storage.
Innovation Solution
A system that uses an AI model trained on a client device to filter group messages based on user-selected faces or text expressions, preventing irrelevant messages from being stored locally and optimizing storage space by only saving relevant content.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If all group multimedia messages are stored locally on the mobile device, then the user has access to all messages, but the device memory is quickly filled and storage space is consumed
Solution Approach 1:
The system performs preliminary filtering of group messages using an AI model trained on user-selected faces and text expressions. This preliminary action occurs before messages are stored locally, preventing irrelevant messages from consuming storage space while ensuring relevant messages are retained for user access.
Solution Approach 2:
The system extracts and identifies relevant messages from the group chat stream based on user-defined criteria (specific faces and text expressions). By taking out only the relevant messages and excluding irrelevant ones, the system optimizes local storage usage while maintaining accessibility to important content.
2Quantity of substance
If manual review and deletion of irrelevant messages is performed, then storage space is freed, but time is spent on manual management
Solution Approach 1:
The system implements self-service filtering through an AI model that automatically identifies and filters relevant messages based on user-selected faces and text expressions. This eliminates the need for manual review and deletion by users, saving time while effectively managing storage space through automated relevance assessment.
Solution Approach 2:
The system uses feedback from user interactions and selected criteria (faces and text expressions) to continuously improve message filtering accuracy. This feedback mechanism enables the AI model to learn user preferences and automatically adjust filtering behavior, reducing manual management time while optimizing storage usage.
3Quantity of substance
If cloud storage is used to store group messages, then storage capacity is increased, but user data privacy may be compromised
Solution Approach 1:
The system extracts only the essential filtering criteria (user-selected faces and text expressions) needed for message relevance determination. By taking out only these necessary elements and processing them locally through the AI model, the system avoids uploading sensitive user data to cloud servers, thereby maintaining data privacy while still achieving effective message filtering and storage optimization.
Solution Approach 2:
The AI model acts as an intermediary that processes messages locally on the user's device using user-defined criteria. This intermediary approach eliminates the need for cloud storage by enabling the system to filter and manage messages locally, thereby increasing storage capacity efficiency without compromising user data privacy.
4Measurement precision
If an AI model is trained on user images, then message filtering accuracy is improved, but processing requirements increase
Solution Approach 1:
The system applies partial action by training the AI model only on user-selected images containing specific faces and text expressions that are relevant to the user's interests. Rather than processing all possible images or data, the system focuses on a curated subset, thereby improving filtering accuracy for relevant messages while minimizing the processing energy required for model training and operation.
Data Source
AI summary
An example system includes a processor to train an artificial intelligence (AI) model on a training set of images on a client device associated with a particular user. Images in the training set of images are selected using face recognition from a digital camera roll on the client device based on a set of faces chosen by the particular user of the client device. All of images in the training set of images include a detected face from the set of chosen faces. The processor is to filter a group message received from a second device based on the trained AI model.